Thursday, November 29, 2012

Disingenuous and Misleading Op-Ed in NYTimes About Prospects for Lawyers

This op-ed, by the Dean of Case Western Law School, which is ranked somewhere in the mid 60s, putting it in the lower half of Tier 1 schools, is grossly disingenuous and self-serving.

First off, let me get right into it by attacking his school directly: Of the 2011 graduates of Case Western Law, only 67% are in positions that require a law degree, and a full 15% are unemployed. This is nearly twice the October 2012 national average of 7.9%.

Additionally, examination of the statistics on Case Western Law's employment page reveals that the vast majority of employed graduates are in small firms, business or government. These are the precise jobs that Mr. Mitchell points out pay $60-80,000 a year. If you graduate with $125,000 in debt, a full $18,000 a year comes out of your post-tax income, and such payments are non-deductible. Now we are talking about the reality that for the lucky ones who do have jobs, they are making from $42-62,000 a year. That just doesn't seem worth it, especially given the long hours, high stress and mundane nature of many of these jobs. Note, importantly, that the average starting salary of college graduates in 2012 was $42,000. So basically, the average graduate of Case Western Law has spent three years and taken on life-changing debt to earn what their younger siblings are making right out of college.

I'd also like to point out that most of Mr. Mitchell's assertions about "average salaries" should be categorically dismissed. Most professions have something like a bell-curve distribution in salaries, meaning that the 'average' salary is actually a meaningful number. However, the curve of salaries in law has a long tail and two distinct peaks: the majority of lawyers will make $50-150,000 throughout their careers, but a very meaningful percentage start at $160,000 a year, before bonus, and wind up making several million dollars a year as partners, often for decades. As a result, the 'average' wage is hideously skewed. Further, as the "white-hairs" retire, more young associates are not being hired to compensate - firms are simply shrinking: according to BLS statistics, by 2020, approximately 360,000 law school grads will be competing for around 73,600 jobs. Super.

Mr. Mitchell's arguments about the value of law school training are also greatly exaggerated: you do not graduate from law-school ready to be a lawyer by any measure. The motto for law school may as well be "come and see if you are good at this": law school functions as a sieve to identify students who are naturally inclined to reading large volumes of text and then regurgitating them in 30-40 page essays written in three hour examinations. These students get As and go to work in big firms. To go into more detail, the 1L curriculum, universally, requires you to read two or three cases per class, sit in a room of 40-100 other students, and the prof picks one student and grills them on details of the case. Then, your entire grade for the semester is determined by that one three hour test at the end. For the vast majority of classes, that's it. I do not see any strong reason to assume this teaching method, or the skills for which law students are given high grades, are either strong predictors of talent for law - a profession that requires a lot of communications savvy, compromise and necessitates comfort with uncertainty - or appropriate methods to nurture talent for law.

Also, keep in mind that using the casebook method to teach law students is the rough equivalent of teaching the history of medicine to medical students. If you are lucky, in law school, you get some real practical experience in a clinic or moot court, or you get to meet some really smart and engaging professors and classmates (disclaimer: I lucked out and had a few fantastic professors and lots of clinical experience). But that's a crapshoot. And, as with professors in any discipline, many law professors are there because of their research or the prestige of the institution, and aren't particularly talented at or inclined to teaching. So whereas medical students are actually practicing medicine, law students, by and large, are discussing the differences between trespass quare clausum fregit and de bonis asperte - a difference that has not been relevant for centuries. Why? How is that remotely useful? Don't take my word for it, however: the NYTimes covered the problem of law school education quite well. It takes years of practice after law-school to become competent as an attorney, a fact newly minted lawyers come to grips with immediately.

It's also worth acknowledging that Mr. Mitchell, educated at Williams College and Columbia Law school, a t-14 school, most likely makes something like $300,000 a year. This may be why Mr. Mitchell argues law school is worth it - for him, it is.

The conclusion of all this is that yes, law is still a viable and profitable career if you go to a t-14 school, or maybe even top 25. Once you start getting past that, however, you will find this same story at every other law school in the nation, and far grimmer at schools rated tier 2 and below. This is why, when friends ask me if they should go to law school, I tell them that they either need to go to a top ranked school or get an extremely generous scholarship to a lower ranked school. Otherwise, it just isn't worth it.

In the end, law schools are vastly over-recruiting and there are too many lawschools, as the ABA has been way too lax in accrediting lawschools. Compare law to other regulated professions, such as medicine, dentistry, or veterinary practice, and you see that in those other industries, admissions to professional schools are actually correlated with market demand. It would be great if the same applied to law schools as well, and, anyway, think about it: who wouldn't want a world with fewer lawyers?

Thursday, September 20, 2012

Some Suggestions For the Times

If it seems like I am critical of the NYtimes, please note that this is because I bear deep affection for that organization, and hope for the best. I know that the entire newspaper industry is in crisis, and I'd rather see venerated and important organizations like the NYTimes make the transition to the digital world (largely) intact, rather than watching them collapse and hoping someone else fills the void.

That having been said, I am very much of the opinion that a simple paywall is not going to get them the money they need to continue to operate, and I'm also of the opinion that online ad revenue will not fill this gap. So, as a result, they need to start selling services that actually represent value above and beyond the basic delivery of news. Additionally, as seen below in other posts, I think that their slow creep towards interest pieces and magazine style focus pieces is really detrimental to their mission. However, that is neither here nor there. Fundamentally, here is a list of things I'd be willing to actually pay a monthly fee for, rather than pay for simple access to news:

1. Access to the writers room. I'd be willing to pay a subscription fee if it meant I can post questions to articles that somewhat reliably get responses from the authors of the various articles. To be specific, I don't need my own questions responded to, but, using a reddit-style upvote system to bring the most common questions to the top would suffice.

2. Access to the op-ed page. The same as above. Individual editorialists vary on their responsivity to the community - both Paul Krugman and David Brooks have published and responded to letters I sent to them - but this should be formally instituted with priority access given to paying customers.

3. Story suggestions. There are a lot of stories that get covered in my sphere of business, specifically, law and technology, that simply are not touched on by the NYTimes. It would be great to have an upvote system where users could submit and vote on stories, and the top ten on any given week would be seriously considered for coverage by the news staff.

4. Sponsored forum events. Everybody loves an AMA, and I know many people would pay good money for the chance to do a livechat, once a week, with a chosen editorialist, writer, or guest.

5. Customized paper delivery. Here's what I'm not willing to pay for: the same printed newspaper received by hundreds of thousands of people that is basically the same content I can get online. What I will pay for is a twenty page, computer assembled newspaper delivered to me daily that has stories that are relevant to me. Thinking this through, many, many people would be willing to pay for this service, which, though clearly technologically tricky to execute, is not infeasible, and would become something of a status symbol very quickly.

6. Ad-free online browsing. I simply do not understand online services that charge you for access but still force ads onto your screen even behind the paywall (I'm looking at you, Hulu). If I was guaranteed ad-free newspaper browsing, not having stories chopped up onto 8 pages, and not having a two inch news column surrounded by 8 inches of ads, I would be far, far more willing to pay for online access. As it stands now, however, I have absolutely no incentive whatsoever to pay twice - once in cash, and once with my eyeballs - for a relatively fungible experience.

The theme across all these suggestions, it should be noted, is acknowledgment that I myself am a human being with my own interests and that I want to interact with the paper, not just read it. The the above features go beyond the mere delivery of news and recognize these two facts. What is certain, however, is that by simply insisting on paywalls for news delivery, the NYTimes will find that it is beating on dead horse.

Monday, September 17, 2012

Uncritical Reporting of Thiel Fellowships in NYTimes

Update: This post was put on Techdirt as of 9.20.12.

 Yesterday, the New York Times had a headline article about Thiel Fellowships.
The program is very interesting, and I was hoping to garner some insight into its success.

However, I was left severely disappointed at the lack of any critical examination of the program, which is still quite controversial. The whole piece read like a PR blast. For instance:

1. The program encourages high achieving indidivuals to skip college in exchange for $100,000 over two years of fellowship grant, plus access to Thiel's network. While many projects discussed in the article were interesting, there was virtually no information about the sustainability of any of these projects or whether or the fellows had achieved any academic success - such as publications - or business success.

2. The article entirely omitted any examination of the fact that Thiel himself has an undergraduate degree from Stanford and a law degree from the same -- this was relegated to a parenthetical. Surely some examination into this seeming contradiction is merited.

3. Similarly, the article totally failed to examine the fact that, at best, this program can really only be successful in a narrow slice of fields -- computer software being the largest. For applied sciences generally, however, and especially engineering and medicine, you simply cannot be an autodidact and a viable career candidate - I'm not allowing someone without an MD to replace my hip, and I'm certainly not allowing someone without engineering qualifications to design my hospital. Too many professional organizations, professional licenses and research areas require formal schooling for this model to be scalable in our most key STEM disciplines. And yes, clearly some fellows are studying applied engineering, such as solar cells, or going into biotech, but there was no critical examination in this article, whatsoever, of whether it is feasible to be an autodidact in these fields which typically require years of graduate level tutelage, even for students well into the genius range. For instance, there was a story about a fellow who was studying gerontology and having great difficulty raising funds - is this, perhaps, because serious VC is not willing to give funding to someone in this field who does not have an MD or equivalent bio degree? I know when raising money for tech startups, VCs frown upon so-called "non-technical" founders - I'd be very surprised if this was not also the case in biotech and electrical engineering.

4. A huge part of Thiel's argument, from what I gathered, is that the network he introduces his fellows to is a large part of the importance of the program. This article seems to be ignoring the fact that, for the vast majority of us, including Thiel himself, these networks are actually formed at institutions of higher learning.

5. The biggest indicator, I think, of the laziness of this article is that the lede is about a student who left Princeton to become a Thiel Fellow, and the "cost" of Princeton was not even the primary factor in this decision. I should certainly say so. Choosing a Princeton student is an exceptionally poor example, as Princeton is renowned for having possibly the best financial aid department of elite American schools, excepting the military academies. Princeton was the first school to completely eliminate student loans as of 2001, and they give extremely generous aid packages. So, from the very get-go the author's credibility was sincerely lessened in my eyes. Additionally, the statement that this article is being written "[a]t a time when the value of a college degree is being called into question..." is highly cynical and glib. What is being questioned is whether colleges are teaching the right things
and whether the current college model is appropriate for all students, or if more alternative forms of higher learning need to be explored. What absolutely no one questions is that college degree holders outperform college dropouts, high school graduates, and especially high school dropouts across a wide spectrum of important metrics, including not only lifetime income, but health and divorce rates. This American Life has an interesting and tangentially related piece on the lasting impact of education on young people this week. Fundamentally, the debate is not whether higher education is unnecessary, it is about how higher education can more fundamentally meet the needs of different types of learners and address growth fields in the economy.

So, New York Times, I'd love to see you take a second shot at this. I'm deeply interested in alternative education methods, however, this article was entirely uncritical. It was a fluff piece that, somehow, made the A1 Sunday Headline online -- which baffles me.

Instead of merely talking about how fantastic the fellows are (I'm sure they are), how exciting the program is (I'm sure it is), or what a fascinating iconoclast Mr. Thiel is (this goes without saying), let's see some critical examination of this program. It has been going on long enough that we should be able to see some verifiable data, comparing fellows' progress to the peers they left behind in school, or, if nothing else, some success stories about gainful employment, fundraising, papers publishing, products brought to market, etc. Instead, this seems like a retweet of corporate PR.

Friday, September 14, 2012

The Trademarkability of Red Soles

A recent decision came down, wherein Louboutin was allowed to keep its trademark on red soled shoes. I think this is a very poor decision, and misunderstands the basic point of trademark law.

Copyright law and patent law (whether or not you agree with their efficacy, overall) are at least explicitly designed to protect creators and their business interests. However, trademark law differs in that its primary purpose is consumer protection: trademark exists to allow for product marking so that consumers can be confident about the origins of products and all their attendant considerations, ranging from quality of product to the reputability of the company supplying said product. For instance, you want to know that the batteries in your remote control won't explode, and if you see the label "duracell" or "energizer" on the battery, you can be reasonably sure that there is a responsible company behind their manufacture. 

However, the consumer is actively harmed by the court's grant of the exclusive right to make shoes with red soles to Louboutin. Not only is it unlikely that the sort of consumers who are going to spend that type of money on shoes are likely to unintentionally buy knockoffs -- we can all reasonably expect that Nordstroms is not stocking its shelves with counterfeit goods -- but when these consumers do buy knockoffs, they know quite well they are doing, and it is an intentional act. However, that is beside the central point, which is the fact that now, if someone who cannot afford to buy a red-soled pump from Louboutin wants to buy a red-soled pump from a competing manufacturer, the competing manufacturer would be committing trademark infringement by supplying said shoe. Which is ridiculous in my eyes. So, as a result, this trademark decision simply favors those with disposable income over those who may be fashion conscious but not made of money, and I simply cannot see any conceivable counterbalancing consumer protection concern.

Here's a simple test when buying luxury brands:

1. While you are holding the item you are purchasing, look around you.
2. Are you standing in a store, the name of which you recognize? Then it is real.
3. Are you standing behind a van or in front of a folding table on the street? Then it is fake.

Trademark protection over color choices in functional design elements is simply unnecessary for brand protection or consumer protection in the world of high fashion. There are plenty other ways to identify fashion brands, such as logos and tags, that accomplish brand identification perfectly well, without having the negative externality of limiting fashion choices for all but the very wealthy. Let me be very clear: I do not believe that competitors should have the right to claim that their own knockoffs are Louboutins - but I don't see a reason why Steve Madden shouldn't be able to make red-soled pumps under his own label as well. Personally, I'd also argue that this is a functional design element, and therefore ineligible for trademark protection as a matter of law - but that is another matter for another post.  

Thursday, September 13, 2012

Fair Use, Gotye and Star Wars

FYI, this is a repost of a post I wrote on my company blog.
Back in August of 2011, my girlfriend sent me a link to a charming music video that had just dropped.  

Within the next year, this video received several hundred million views, and propelled two young artists -- Gotye and Kimbra -- to international fame. Clearly, this was a huge win for them, and I'm sure this video generated tons of additional ticket, album and t-shirt sales, despite the fact that they probably made peanuts from the video itself. This video was a loss leader, driving traffic to paid services, while itself representing a largely sunk cost.

The video became a cultural phenomenon. And, as the saying goes, the sincerest form of flattery is imitation. Not surprisingly, some extremely well done remakes were created, which also received millions of views on YouTube. For instance, here is the talented band "Walk off the Earth" performing the song on a single guitar being played by five people. 

It’s pretty awesome, and has received more than 133 million views as of this pub date, making it one of the most successful YouTube videos of all time.

Here is a version of the song reworked about Star Wars, a video that I personally find to be even more moving than the original (Han shot first.) 

I'm sure many people my age agree, and it’s already reached nearly five million views.

You can react two ways to the creation of these videos: the first is to recognize that they are flattering forms of tribute and bring attention, and therefore additional opportunities for revenue to the original video itself. The second is to send DMCA (Digital Millenium Copyright Act) takedown notices to YouTube claiming that these videos infringe the copyrights of the& original artists.

Luckily, it appears that Gotye and Kimbra are perfectly happy to take the first approach, and, collectively, we should thank them for that. Technically, they would be within their rights to send takedown notices because these are clearly derivative works within the meaning of Title 17, the US Copyright code. An example of an artist taking the second

path is illustrated by the now notorious Lenz case wherein the Artist Formerly Known and Now Again Known as Prince sent YouTube a DMCA takedown for a video in which a child was dancing and, in the background, a Prince song was playing. Years of litigation and substantial cost was incurred, resulting in the court determination that a rights holder must consider fair use before sending a takedown notice.

The problem, however, is that very often, such considerations are just not made. Automated systems generate DMCA takedown notices and send them out en masse, but fair use determinations, by definition, need to be made by human beings using their reasonable judgment.

Now, there is a strong argument that both of the above videos are, in fact, fair use. However, the way that our copyright system is set up requires that YouTube would need to take down the videos upon notice from the original rightsholder, as fair use is a so-called "affirmative defense." This means that in order to make a claim of fair use, you need to hire a lawyer to write a response letter, and the default judgment is heavily in favor of the original rightsholder.

As a result, we have a system in which the first person to put up a new video on YouTube has the de facto right to send DMCA takedown notices to any imitators, even if there is no demonstrable harm caused to the original artist by the publication of the imitations. In other words, the fair use exception is a check built into copyright law out of respect to the fact that certain types of re-use and commentary on copyrighted works are necessary for society due of free speech considerations, and also, there is no actual harm done to the artists by the imitation of their works. Yet, because of the DMCA takedown process instituted in 1998, the default system now actively harms imitators, who need to have lawyers on retainer to send response notices to YouTube in order to have their own videos put back up. Personally, given that I'm of the opinion that most great art includes a remix of one kind or another, I'm not happy that our copyright system is predicated on the idea that imitation is harmful.

Because, honestly, I like the Star Wars version of the song better.

Thursday, August 16, 2012

Groupon is Not A Tech Company

Tl;dr A tech company is a company who's business model centrally relies on addressing a problem with a unique or novel technological solutions. Groupon's use of technology is incidental to its business model. 

I was reading this article today in Wired about how Groupon's stock fall-off is sharper than other "tech giants." While that is interesting, I think it is incorrect to call Groupon a tech company, and, accordingly, I don't know that it is quite fair to compare Groupon to tech companies. This is in no way a judgment on the quality of Groupons services, the dedication or talent of their team, etc. They executed well and have many very satisfied customers; that alone is praiseworthy. However, Groupon simply is not in the tech business: They sell coupons, en masse. That's it. Doing so did not require any technological innovations, either in software or hardware, and that is why their model is infinitely replicable. This, in fact, it its primary weakness: Groupon simply has no barriers to perfect competition. And when I say 'perfect competition' I mean a competing product offering that is 100% satisfactory and 100% fungible. This is not to imply that technological innovation is necessary to create such barriers: many, many non-technology based services do have such barriers, ranging from businesses that rely on artistry, editorial content, licensing, to networking and access to supply chain, etc.

Importantly, the simple fact that a primary sales interface is web-based doesn't make a company a tech company any more than selling products in brick and mortar stores would disqualify a company from being a tech company. Let's compare to Amazon.

Amazon, at its inception, sold physical books over the internet -- at first glance, this does not appear to be a 'tech' company. However, Amazon was confronted with a variety of relatively new engineering problems regarding the scalability of their website, online sales and computerized supply chain management. So, despite the fact they sold books, they heavily utilized technology to do so, and one of their main competitive advantages was their leveraging of technology to create this massive scalability. Since then, Amazon has firmly cemented itself as tech company, as Amazon Web Services powers much of Web 2.0 and the Kindle fire is a truly revolutionary product.

So, in comparison, Groupon has not apparently solved any problems with novel applications of technology. Neither did

Importantly, Groupon's primary advantage is that it was the first mover, and this advantage will not last long. And it's for that reason that I think that and Groupon will share a similar fate. Anyone can sell kibble on the internet, and anyone can offer coupon deals online. In fact, quite in contrast to Groupon's model, which is based on employing warehouses full of comedians to write haikus about discount massages, many companies are popping up left and right that do leverage technological solutions to offering online couponing that are scalable at nearly zero marginal cost. Groupon's model, however, is not scalable, and probably, is not sustainable either. I wish them all the best, but I truly believe that unless they start innovating technologically, they will soon lose their first mover advantage. I believe that there are enough talented people at Groupon to make the pivot, but whether they will pivot or simply try to double down on being the first to market is yet to be seen.

Friday, June 22, 2012

Tim Wu: Computers Don't Inherit Their Programmers' Free Speech Rights. But Why Shouldn't They?

Tl;dr In an earlier post, I argued that it is important to recognize that search results should be treated more like editorial content than as mere assertions of fact. Titan of Internet Law, Prof. Tim Wu of Columbia Law, frames the argument differently, and though I think we fundamentally agree, it is worth clarifying why I find his characterization somewhat troublesome. Summary: It is more important to subject automated speech to First Amendment analysis, even if weakened as the result of considerations due to its commercial and automated nature, than to categorically exempt it from procedural constitutional protection. 

Edit/Update: Techdirt has weighed in, and it is gratifying to see that a lot of the same arguments I made below are made by Mr. Masnick and the scholars he references.

Well, I'm stepping right into the line of fire on this one. Prof. Wu is pretty brilliant, I often agree with him, I find him deeply insightful, and, basically, for me to disagree with him is kind of like an ant pulling on Superman's cape. That aside, I do disagree with Prof. Wu's position on this issue, and I think that some law actually backs up my point. For the purposes of this post, I'll call the speech in question -- results of search engines, automated ranking algorithms, stock ratings, etc. -- automated speech.

To summarize Prof. Wu's argument, it appears to be something like the following:
• 1. Automated computer output, by default, should not receive full First Amendment Protection;
• 2. It doesn't make sense to say that automated output should 'inherit' the constitutional rights of the authors of the automated processes;
• 3. It would be troublesome to see First Amendment defenses to Anti-trust arguments.

So I'll make three points in response.

1. Analysis of 'automated speech' should begin with a determination as to whether it qualifies for speech. IT should not assume that such speech fails to qualify as speech by default. Established law demonstrates that many types of automated speech qualify for copyright protection, and it is very confusing to assert that something that is the subject matter of copyright should not be treated as speech.

• First off, Google's ranking database should qualify as a compilation, as defined by 17 USC 103(a). This alone should merit copyright protection.
• Second, in both CCC Information Services, Inc. v. Maclean Hunter Market Reports, Inc. 44 F.3rd 61 (2nd Cir. 1994) and CDN Inc. v. Kapes, 197 F.3d 1256 (9th Cir. 1999) it is clear that compilations which are based on the expert judgment of facts -- in the former case, predictions of car prices, in the latter, coin prices -- are sufficiently creative to merit copyright protection.
• In my opinion, Google is therefore not only the subject of copyright protection as it is a compilation, but, based upon the above case law, it is also creative, as it is the result of expert judgment. Both the above cases are silent on the issue of whether judgment can be the result of an algorithmic process, and, in my eyes, it is irrelevant. (If anyone has any contradictory authority, please send along!)

This leads us to a confusing position, wherein Prof. Wu is arguing that things that qualify for copyright protection should not be treated as speech by default -- I think his assertion is that they need to be 'justified' as speech. I am arguing the inverse, that the default rule should be an assumption that automated speech is speech, and then the burden would be on the challenging party to prove otherwise.

I could be wrong on this, but, from what I understand, while there are many actions that count as "speech" that are outside of copyright, there are no copyright rights that are outside of the realm of speech. The subject matter of copyright, for reference is defined as:
(a) Copyright protection subsists, in accordance with this title, in original works of authorship fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device. Works of authorship include the following categories:
(1) literary works;
(2) musical works, including any accompanying words;
(3) dramatic works, including any accompanying music;
(4) pantomimes and choreographic works;
(5) pictorial, graphic, and sculptural works;
(6) motion pictures and other audiovisual works;
(7) sound recordings; and
(8) architectural works.
(b) In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.
I just cannot conceive of a hypothetical situation wherein it would be the case that something that is copyrightable is not speech, but I may well be wrong. Either way, it makes me deeply uncomfortable to think that there could be a situation wherein a copyrighted work is categorically not speech -- perhaps I can come up with a really elaborate architectural example, but it would necessarily rest on the total separation of the form of an architectural work from any artistic, cultural or social commentary whatsoever. That is deeply unsettling.

2. The assertion that the output of automated processes designed by humans, exercising their first amendment rights, should not receive the same procedural analysis as outputs of those humans themselves, seems to be an arbitrary distinction. 

I'd argue that Wu's reasoning:
Defenders of Google’s position have argued that since humans programmed the computers that are “speaking,” the computers have speech rights as if by digital inheritance. But the fact that a programmer has the First Amendment right to program pretty much anything he likes doesn’t mean his creation is thereby endowed with his constitutional rights. Doctor Frankenstein’s monster could walk and talk, but that didn’t qualify him to vote in the doctor’s place.
Is conclusory. It doesn't state why this should be, just that it is. Personally, I just do not see why the act of a human applying each set of transformations to a set of input data magically imbues it with 'speech-hood,' as opposed to a computer performing this process by proxy. It is perfectly reasonable to me, for instance, that the creativity necessary to reach the bar of speech is met in the creation of a ranking algorithm and the selection process for its data inputs. Let's put this another way: a regulation covering algorithms that produce publicly viewable results is directly a regulation on the freedom of speech of the programmers that programmed the algorithm, and, additionally, such a regulation is unlikely to produce predictable results, much less results desired by legislators or consumers. I simply cannot see a way around the free speech analysis as a result.

It appears to me that Prof. Wu is basically asserting that when you have an intermediary process between human thought and output, you break the status of speech-hood. This doesn't sit well with me, as we rely on increasingly on technology to express speech, and, importantly, art.

For instance, I use photoshop a lot, and it produces a tremendous variety of effects that I would in no way be able to create with my own hands -- yet my photoshopped images receive full First Amendment protection, despite the fact I have used a variety of totally automated effects and algorithmically based filters in their execution. The same applies for audio editing software and video editing software.

I think Prof. Wu is held up on the semantic nature of text, but I think, increasingly, as NLP becomes more sophisticated, this line will be increasingly hard to draw. I currently have, in the "Services" menu of my Application menu on my Mac, a choice called "Summarize." It uses an algorithm to create synopses of text I have highlighted. Should this automated summary not receive protection as speech? What about the contrapositive? It will not be long before I can feed shorthand notes into MSWord and it will produce full english text, or at least a close enough approximation that it would fool an 8th grader. Should this not receive protection either? A colleague of mine has pointed out that he feels Prof. Wu is hung-up on this because automated drafting of this kind is currently rudimentary, but it will increase in sophistication rapidly over the coming years. I agree.

The point is, I think that the distinction Prof. Wu is insisting on is artificial, and doesn't accurately map to real world use or user expectation.

3. Google has not asserted a First Amendment defense against anti-trust violations, but that is due to the reasons that there are no good anti-trust claims and other defenses are available.
To Google’s credit, while it has claimed First Amendment rights for its search results, it has never formally asserted that it has the constitutional right to ignore privacy or antitrust laws.
I won't reproduce the whole argument here, I'll just refer again to my previous blogpost. Suffice it to say, and, again, I'm not an expert in this topic, but I feel the anti-trust arguments against Google are pretty thin.

To give computers the rights intended for humans is to elevate our machines above ourselves. ...
And that’s where theory hits reality. Consider that Google has attracted attention from both antitrust and consumer protection officials after accusations that it has used its dominance in search to hinder competitors and in some instances has not made clear the line between advertisement and results. Consider that the “decisions” made by Facebook’s computers may involve widely sharing your private information; or that the recommendations made by online markets like Amazon could one day serve as a means for disadvantaging competing publishers. Ordinarily, such practices could violate laws meant to protect consumers. But if we call computerized decisions “speech,” the judiciary must consider these laws as potential censorship, making the First Amendment, for these companies, a formidable anti-regulatory tool.

That top comment is the one that gets me: it's nonsense and it is fear-mongery. As Prof. Wu himself stated many times, a point I agree with, a point that is one of the foundational pillars of my argument, is that the analysis must start with First Amendment considerations, but may well, and probably should, wind up determining that automated speech has less protection than core speech because of the precise reasons that Prof. Wu has cited. No one is saying that automated speech should have the same status as core speech, and absolutely no one is saying that machines should have more rights than humans. All I am saying is that the analysis needs to start at the same place, where Prof. Wu seems to be suggested it can be circumvented altogether.

The issue that I believe Prof. Wu is getting at is that if we assume that automated speech is, in fact, speech, it makes it harder to regulate. So? It should be hard to regulate, and as Prof. Wu has himself asserted, because automated speech is often commercial in nature, and compounded with the fact that it is, in fact, automated, it is entitled to less First Amendment protection than core, political speech by a human. I am okay with this. 

Additionally,  I can see virtually no benefit from making it easier to regulate automated speech: legal attempts to regulate code are often ham-fisted, sometimes disastrous, and, at best, extraordinarily difficult to craft well, especially at the hands of a luddite legislature. For instance, I can see no reason to assume that lawmakers telling Google how to mess with its ranking system would in any way benefit consumers -- Google's smartest competitors have failed to come up with a system as effective as Google. Why should we assume a bunch of lawyers on Capitol Hill have anything meaningful to say about their algorithm?

If legislators want to regulate automated speech, the proposed regulations should have to face the rigor of First Amendment analysis rather than circumvent such analysis by default simply because there is an automated intermediary between the author and the content. If it turns out that automated speech deserves less protection than other types of speech, this should be decision based on First Amendment analysis, not based on skipping the analysis completely.

Tuesday, June 19, 2012

Charles Carreon: Plot Thickens as Internet Collectively Realizes He Has a Pretty Serious Disciplinary History

Tl;dr Lawyer Charles Carreon, representing himself pro se against theOatmeal's Matt Inman, the American Cancer Society and National Wildlife Foundation (good luck with that), apparently has a pretty serious history of being censured by the courts for misconduct. I therefore nominate him Jack Thompson 2.0.

According to Wikipedia:
In October 2005, Carreon was suspended by the Oregon State Bar for 60 days for the unlawful practice of law and failing to deposit or maintain client funds in trust.[11] In September 2006, Carreon was also suspended for two years by the State Bar of California, stayed, and placed on two years of probation with an actual 60-day suspension for violating his duty to maintain client funds in trust, and for practicing without a license in Canada.[12]
Please note that these are not trifling matters. It is drilled into your head constantly in law school that mishandling client funds is just about the worst thing a lawyer can do, short of lying to a tribunal or straight up committing felonies. It is understandable that one could mess up... but twice, for the same charge? This is less excusable, particularly in the added presence of a censure for practicing without a license in not one, but two jurisdictions. I think the top comments on reddit pretty accurately summarize the situation:
callipygian1 144 points 1 hour ago
as a retired lawyer, i know we have many assh*les in our profession, but damn few wikipedia-grade assh*les.

BaconCat 30 points 50 minutes ago
I propose a new classing system for assh*les, with wikipedia-grade assh*les being the top tier.

At this point, it is safe to say that the internet has officially labeled Charles Carreon a "Wikipedia Grade Assh*le." Can this somehow make it into his Wikipedia entry?

What do Spoilers Spoil? The Fun.

Tl;dr Stanley Fish, consistent with his character, has a rant castigating the phenomenon of 'spoiler alerts' - it is simultaneously condescending, clueless and socially inept. Spoiler alerts are not about messing up the experience of seeing a movie or reading a book, they are about social grace and communal experience.
To summarize his argument, Stanley Fish has simultaneously asserted that:
1. Actually alerting readers/viewers to spoilers is beneath his dignity;
2. People who get upset at a lack of spoiler alerts are actually incorrect, whether they know it or not, because some spurious 'science' that he cites says that they actually like having plots spoiled;
3. Any work that can be 'spoiled' is trash.
I disagree.

I harbor a rather strong dislike for Stanley Fish's NYTimes column. First off, he is a professor of law who did not go to law school, and, usually, when I read his columns about law, I find that this fact shows. Additionally, I find his opinions regarding education to be consistently condescending, and ranging from cluelessly elitist to pridefully elitist.

Edit/Update: Mr. Fish's recent post about hate speech is actually rather well measured. I don't agree with all of it, but it is thoughtful and even-handed, and a worthwhile read. Please note I'm not saying Mr. Fish isn't a talented writer, or that his opinions universally lack merit: he often has very insightful stuff to say. However, I often find his tone very to be dismissive, and his approach to education, I believe, is a very good example of a mindset common amongst administrators and educators that contributed directly to the current crisis of higher education in my generation. The crisis is manifest both in terms of unemployment among my generation -- and though this is obviously due in large part to the greater economic situation of the country, it is also due to a lack of practical, applied education and useful skillsets -- and massive debt -- as it appears that Fish holds education in such a privileged position that cost of education is not a factor that is even up for consideration: if you get accepted, you go, end of story. These two positions, taken in tandem, are guaranteed to produce a large swath of debt ridden students with no practical skills, and I think he insists that, somehow, this is okay or 'worth it' due to the intrinsic value of liberal arts education. That is why I call him both condescending and clueless -- not because I think he is personally a jerk (though I call him one for comedic effect), but, because I feel he is callous to the crisis afflicting my generation. Oh, also, I say mean things because they are funny. Typically, however, I only do this in my writing for pleasure: my professional work product is dry and technical, and this blog is my sanity-vent. Back to the regularly scheduled blog.... End edit.

His latest column (and I admit I'm late in penning this response) is no exception. I'll summarize his argument: some recent studies indicate that viewer/reader suspense is heightened if one knows the spoilers of the plot in question. Personally, the way these studies are explained makes me question their validity. For instance:
Scholars have come up with three ways of either de-paradoxing the paradox [ed note: the paradox in question is the heightened sense of suspense despite knowing the spoilers] or denying it. Robert J. Yanal argues that repeaters mis-describe their own emotions; they might feel apprehension or fear in relation to a foreknown event, but they mistakenly report it as suspense: “apprehensiveness is not suspense, though the two often occur together” (“The Paradox of Suspense,” British Journal of Aesthetics, 1996). In short, there is no paradox of suspense.
Consistent with this, the gravamen of Fish's argument is basically that if someone thinks they enjoy spoilers, they are wrong, and science proves it.


I don't particularly want to engage with the merits of that argument. Like I said, I think the 'science' above, is questionable. What I want to engage with, however, is the following, brilliantly detailed by his first and last paragraphs:
Over 10 percent of the comments on my “Hunger Games” column brought up the question of spoiler alerts. “Haven’t you heard of a spoiler alert?”, one exasperated reader asked. Another reader, Jim, reported that he was “trying rapidly to withdraw my forward of the article to my wife who’s in the midst of the 2nd book.” He didn’t want his wife’s experience spoiled as it would be, he assumed, if she knew how things turned out. ...

In either case, the spoiler doesn’t amount to much and alerting readers to it is not a high obligation. If “The Hunger Games” is so shallow that it can be spoiled by a plot revelation, the alert doesn’t save much. If “The Hunger Games” is a serious accomplishment, no plot revelation can spoil it.
I think it is safe to say that Mr. Fish is someone that you would simply not want over to dinner, unless you were serving dinner on top of him, because he has the social understanding of a table. I'll explain. [ed note: needlessly harsh, but I think it's funny, so it stays in.]

The allure of 'spoilers' is not only the adrenaline rush of the discovery, which, despite Mr. Fish's insistence otherwise, is a valid sensation, but, importantly the joy and thrill of a 'spoiler' is the shared experience of discovery. It is not an accident that people go out to movies in groups, or that the shared experience of television is what drives ratings, or that talks "at the water cooler" are strong indicia of cultural phenomena. In other words, spoilers are not simply a personal issue, they are about the interpersonal relationships formed around mutually experienced dramatic events. I can only assume Mr. Fish has never seen a movie, play or television show with his family or friends. Either that or they have left in disgust as he explained that there is no Island, Bruce Willis is dead and actually, the Planet of the Apes is Earth. [ed note: terrible movies chosen intentionally.]

Additionally, with regard to the second paragraph, quoted above, Fish is really being a hypocrite. He constantly exhorts people to read and enjoy the liberal arts for their own sake, yet he is implicitly denigrating a series of books that have actually persuaded millions of Americans to read. To put it another way: it is not a sound strategy, when trying to convince people that reading is a worthwhile and noble past-time, to simultaneously assert that books that excite millions of people -- the books driven by spoilers, like Da Vinci Code, everything written by Michael Crichton, half of everything written by Stephen King, the list goes on -- are shallow. I can only assume, that, again, consistent with his character, Fish wishes people were reading books that he personally approves of.

Please be assured, next time Mr. Fish opens his mouth, I'll be there to stick my foot it in.

Monday, June 18, 2012

How to Fail at Internet Lawyering: Charles Carreon v. The Oatmeal / The Internet / Common Sense

Tl;dr Internet Lawyer Charles Carreon is making a really big fool out of himself, after he sued on really spurious grounds, and is now suing the charities that made donations to as a response to the original lawsuit. The situation is totally pathetic, and Carreon is really actively destroying any goodwill he may have had with the netizens of the world.

The fine folks at Techdirt have done an excellent job of reporting on this story. Spoiler: if you are going to advertise yourself at being an internet lawyer, it doesn't help to be clueless at the internet.
If you've been away from the internet for the past week, this story started as an online dispute between Matthew Inman, creator of the webcomic The Oatmeal, and a site called Funnyjunk, which lets users post content to the site. About a year ago, Inman wrote a blog post complaining about Funnyjunk's reposting of his webcomics. As we've noted a few times, Inman's statements about Funnyjunk were overly aggressive -- and did mention "stealing" of his own work. He seemed to ignore that it was users who uploaded the content. However, while we disagree with Inman's original characterization of Funnyjunk and how it operates, it certainly did not reach the level of "defamation." Also, we appreciate that Inman chose not to sue, but rather to make use of the court of public opinion. In response, Funnyjunk lashed out, incorrectly stated that The Oatmeal wanted to sue him (when Inman very clearly stated he had no intention to sue) and also asked a bunch of Funnyjunk users to contact Inman.

Everything seemed to die down, until about a week ago, when lawyer Charles Carreon, representing Funnyjunk, sent a letter to Inman, threatening to sue Inman for the initial blog post, claiming that it was defamation and a Lanham Act (trademark) violation for false advertising. Neither claim makes much sense, and Inman responded with both an excellent letter from (occasional Techdirt contributor) Venkat Balasubramani, and Inman's now famous annotated letter, leading to an IndieGoGo campaign to raise $20,000 (the amount Carreon/Funnyjunk demanded from Inman) for two charities: The National Wildlife Federation & the American Cancer Society.

Following that, Carreon told MSNBC he intended to shut down the fundraisers, and then bizarrely accused Inman of "instigating security attacks" against his website. Finally, on Friday he told Forbes that he wasn't backing down and that there had to be "something" in the California code that he could sue Inman over.

Apparently he's found something. As reported by both Ken at Popehat and Kevin at Lowering the Bar, Courthouse News has a notice saying that Carreon has filed a lawsuit in the federal district of Northern California. And, as mentioned above, he doesn't just sue Inman, but also IndieGoGo and the two charities. Yes, the two charities. I'll repeat that again: Charles Carreon appears to be suing two of the most well known charities because Matthew Inman asked people to donate to them.

It gets better. Read on here. Please, click that link and send some traffic Techdirt's way. And if you don't read them regularly, you should. Update: Also, the response from TheOatmeal's Lawyer is really, really good.

Saturday, June 16, 2012

The .gif is 25 Years Old, So Let's All Make Sure to Pronounce it Right: JIF

Tl;dr It's pronounced 'jif' not 'gif' -- end of story. So the gif is 25 years old as of June 15th, 2012.

Let's all take a moment to recognize the correct way to pronounce 'gif', right from the horse's mouth:
It's pronounced like "jif". Period. The end. That's final. End of story.

You disagree? Hey, I'm just quoting the inventors of the format. Here's the evidence:

CompuServe used to distribute a graphics display program called CompuShow. In the documentation for version 8.33 in the FAQ section, it states:

The GIF (Graphics Interchange Format), pronounced "JIF", was designed by CompuServe and the official specification released in June of 1987.

There, straight from the inventors of the format.

Convinced yet?
Also, on a personal note: gif, I love you.


Enjoy, and, .gif, happy birthday.

Is It Just Me, or is the iTunes Store Throttling its Bandwidth?

Tl;dr I may be entirely wrong, but it appears iTunes store may be throttling bandwidth. Anyone know the full story on this?

Here are two images demonstrating my current concern:

To summarize: despite having decent bandwidth, a 47.4 megabyte mp3 takes 15 minutes to download. However, my math indicates that, with that bandwidth, it should take me just under two minutes to download a file of that size.


That is one minute forty two seconds. Accordingly, it seems that iTunes is operating at about 11.5% efficiency, or throttling to the tune of nearly 90%.

How's my math?

Wednesday, June 13, 2012

WSJ Very Sincerely Doesn't Understand Bandwidth Or Monitor Resolution

Tl;dr The WSJ performs some truly sloppy reporting when it states that the new MacBook Pro, with Retina display, will bring corporate networks to a halt, as they require more bandwidth.

The Wall Street Journal, today, had this to say:
CIOs with loose bring-your-own-device policies might find their corporate networks clogged should employees bring the just-announced Macbook Pro computers to work. Introduced at Apple’s developer conference Monday, the new Macbook Pro is fitted with a Retina display, whose resolution of 2880-by-1800 pixels packed into a 15.4-inch screen is the crispest screen for a computer yet, clearer than Apple’s newest iPad.

But it may also wreak havoc on CIOs’ networks and connectivity budgets — better quality displays require more network bandwidth, which allows users to increase data consumption. Consider that experts told CIO Journal earlier this year that the new iPad, which includes a Retina display of 2048-by-1536 resolution with 3.1 million pixels, would slow enterprise networks to a crawl and increase data costs from carriers. Now imagine how a Macbook with 5.1 million pixels — two million more than the new iPad — will increase data traffic in office networks.
Notice the bolded text. The technical term for this statement is "poppycock."

While it is true that if websites start to offer higher resolution images for consumption on retina displays, this may increase bandwidth demands, the simple act of hooking up a large monitor to your network has precisely zero impact on your bandwidth requirement. None, nadda, {Ø}. If you increase monitor resolution, but the website you are viewing does not serve larger images, the website simply appears smaller on your monitor, or you decrease your monitor resolution to compensate. Increasing your monitor resolution does not magically 'add pixels' to the images you download from the internet.

The above assertion by the WSJ is so flagrantly incorrect that even bothering to explaining why it is wrong is deeply frustrating. One may as well assert, as one redditor put it:
More pixels = more bandwidth It's similar to why my Ipod weighs a ton after I fill it with music.

One may argue that, as more websites seek to offer rich media experiences that are native to the retina display, bandwidth consumption will increase... but again, this has nothing to do with the settings on your local machine (one can argue that websites may auto-detect such settings, but then just instruct your employees to lower their monitor resolution /thread). In fact, at work, I have two 21" wide screen displays, both at 1920x1080, for a total of 4,915,200 pixels. Which is basically the same as the above Macbook. Nearly everyone at my office has this same configuration, and our network is utterly unaffected by this.

No matter what resolution I set my local computer to function on, the internet still serves up the same connection. You can test this, right now: go to your system preferences, and mess around with your display resolution. Then go here: Spoiler: resolution changes nothing. Okay, I've kicked this horse enough.

Suffice it to say, this is just pathetic reporting on the part of WSJ and hopefully will get a retraction and/or correction. It's absurd nonsense.

Wednesday, June 6, 2012

Passwords in Wallets? Maybe Not So Much...

Bruce Schneier wrote, a few years back, that a good idea for password safety is to write down your passwords and put them in your wallet. I brought this up at work and a coworker immediately pointed out that this would be disastrous if you ever lost your wallet: every form of identification, giving away, among other things, where your bank accounts are, are now in the same location with passwords to those same accounts.

So I have to say, after thinking this through, I don't agree with Mr. Schneier on this one. I wonder if he has also changed his opinion on the matter?

KPCB Internet Trends: Ad Dollars Shifting from Print to Online

Tl;dr Ad dollars aren't evaporating, they are just following eyeballs from off-line to on-line.

Mary Meeker's 2012 "Internet Trends" presentation also serves as the basis for this post -- there may be a few more to come. It really is a great presentation, and worth going through. Anyhoo...

I've talked in some recent posts comparing the value of eyeballs on traditional media to new media.

Below are a few screenshots from Ms. Meeker's presentation, that I think further support my overall point (or, maybe, I actually support hers? Kind of hard for the completely obscure anonymous netizen to claim support from a titan of online industry...)

I think these graphs pretty strongly show that the money is not disappearing, it is just being redistributed. Money is moving from off-line to on-line, and a lot of organizations (read: newspapers, television, radio, movies) are worried that they will not be able to make enough money, or any money, online. I simply do not think the numbers support this proposition, above.

I still argue that online ads, when properly done, can be more valuable than offline ads, because they can actually be targeted to consumers. Of course, I also say this as someone who very rarely clicks on online ads. But then again, I also never really paid attention to TV ads, so I don't think I'm the target consumer. In any event, online advertising provides many more meaningful ways to actually connect with users than offline advertising ever did, and, as it evolves, traditional media outlets will be able to take increasing advantage of this phenomenon. So the money isn't going away, it's just moving around.

KPCB "Internet Trends" Presentation Shows Netizens Happy with Search; EU Commission Still Threating Google with Antitrust Violations Over Its Search Methodology

Tl;dr Netizens appear to be satisfied with the quality and usefulness of search results, yet the EU Commission still insists that Google may need to change search results out of anticompetitive fears. This likely means businesses that are upset about their poor rankings, but this is a morally poor basis (though, perhaps, a legally valid one, time shall tell) for an antitrust action.

Over at Techdirt, Masnick has made several good points about the poor arguments put forth by EU Commission in their current antitrust investigations into Google.
There's been plenty of buzz about the EU Commission basically admitting that it's going to go after Google on antitrust grounds if Google doesn't change certain practices. However, as we've seen with past claims of antitrust issues with Google, when you begin to unpack the complaints, they don't seem to hold up to much scrutiny. There is no argument that Google is a big player, and could be abusive -- but there is little, if any, evidence of actual abuse. And, most of all, there seems to be no evidence that Google's actions harm consumers.
However, Kleiner Perkins partner Mary Meeker's 2012 "Internet Trends" presentation has pretty persuasive data against this position. Specifically, the screenshots below, which show that internet users overwhelmingly trust the results of search engines, find them unbiased, and, importantly, that they are becoming more relevant and useful over time:

So, in other words, it is simultaneously very hard to argue that there is harm to consumers as a result of Google's search returns, especially given that the above stats are for search engines overall. Therefore, either (a) there is enough competition in the field that non-Google search engines give users what they want without being dominated by Google (i.e., no consumer harm because of Google's actions), (b) Google itself is providing results netizens find trustworthy and increasingly useful (same) or (c) a combination of both (duh).

Unfortunately, as Wikipedia points out, Anti-trust does not have its basis in consumer protection:
Efficiency-oriented economists reject the goal of competition and instead argue that antitrust legislation should be changed to primarily benefit consumers. No Congress or administration has supported this position. These economists largely ignore the political issues that motivated the laws in the first place.
Fundamentally, I totally disagree with this point of view. Antitrust should be about consumer protection. I honestly cannot think of a circumstance where considerations purely of business competition, that do not in any way wind up impacting consumer protection, should be the basis for trust-busting. The only potential hypothetical that comes to mind is a big player ruthlessly squeezing tiny competitors out of a market by undercutting prices... but I don't see why:

1. This is necessarily a bad thing, as I honestly do not see the problem with a business undercutting its competitor and bearing losses a competitor cannot for a period of time as a way to establish market dominance;

2. If there is a concern here, it is because entrenched player will then pump up the price after the competition goes out of business. This is a problem for consumers, and the combination of those two things is definitely problematic in my point of view.

Anyway, in a post that will shortly follow, I'll examine this issue from the perspective of American Copyright law covering databases. Suffice it to say, I simply don't think that Google rankings count as facts, nor are they presented as facts, but are actually purely creative opinions. As a result, I do not see why Google's search activities should properly fall under any sort of unfair competition law: what Google is doing is publishing speech that consists of its opinions on the search worthiness of links. Fundamentally, then, I think free speech concerns simply trump the business concerns in question.

Tuesday, May 22, 2012

Enough Facebook, Anyone Remember AOL?

Tl;dr AOL is operating at near FB levels of P/E, on tiny earnings, and this is probably unsustainble.

Tim Armstrong, CEO of AOL, did a talk at Techcrunch Disrupt. In it, Mr. Armstrong said, inter alia, that AOL was “becoming a house of strong brands” as it transitions to be more of a content company:

He admitted that despite having some dissident shareholders that “don’t believe”, in the content strategy, most of AOL’s shareholders do believe in it.
I'm not particularly interested in Mr. Armstrong's claims about what AOL is doing. However, I would like to examine the financials of AOL, especially now that we are all in stock-markety-valuation-y moods given FB's bump. Here's AOL's chart, as of close today, 5/22/12:
Just to reiterate, that is a Mkt Cap of 2.54B and PE of 85.38 which translates to earnings of $29.75M. That is very sincerely not a whole lot of money. Again, assuming a 'healthy' PE is somewhere around 13, AOL needs to post earnings of around $195M, or roughly a 650% increase in earnings. This seems... unlikely, to put it mildly.

So, Mr. Armstrong, it may well be that the shareholders do believe in the current strategy... but they may well be delusional. A PE of 85 on earnings of $30M is just unsustainable for a company with 5,660 employees, and I just do not see that AOL is going to be the company to solve the content-monetization crisis. Incredible to believe that AOL Time Warner was once a hundred billion dollar company.

From Techcrunch: How the Media is Wrong About Facebook's IPO -- But I'm Still Bearish

Tl;dr Though the FB IPO may have gone of a lot more smoothly than many critics are saying, FB needs to post consistent, high growth in order to justify its valuation, which looks unlikely.

Techcrunch contributor and VC member Dan Scholnick has makes three very astute points about the Facebook IPO, worth reiterating:

1. The best IPOs maximize capital raised by the company while minimizing dilution for existing shareholders and employees.
2. The best IPOs minimize the fees paid and value transferred to third parties.
3. The best investments are determined over months and years, not hours and days.

In general, I agree. And by these metrics, FB's IPO was pretty good. Additionally, now that NASDAQ:FB has closed at precisely 31, as of 5/22/12, I think one could argue that the underlying value of FB is actually pretty well within the range predicted by bankers. That is, it seems counterintuitive to me that one would price a stock as lower than its value when planning an IPO, especially as Mr. Scholnick points out, the 'pop' associated with such undervaluations mostly benefits bankers and investors, and not the company. To put it another way, the actual price of FB, purely as a function of what the market will bear, turns out to be pretty close to its IPO, so that seems pretty good. If it has been way undervalued or way overvalued, this would be a problem, as it means the bankers would have improperly estimated the value of the company. So, by the above metrics, and judging by the price the market will bear, FB's IPO seems to have gone off relatively smoothly. However, all this aside, I am still bearish on FB, for the following reasons.

FB's PE is now 99.22, with a Mkt Cap of 66.28B, translating to earnings of 668M. Assuming that a 'healthy' PE would be somewhere around 13, this means that investors are expecting FB to be able to post earnings of 5.01B within the not-too-distant future.

On the one hand, given that FB has 900M users, that comes out to a total user-year value of about $5.66 a year. Given that it looks like FB is on track for something like 80 hours per user, per year it doesn't seem unreasonable to think that FB could get five bucks and five bits out of 80 hours of user usage. On the other hand, this would also represent a 750% increase in revenue, which is, as they say, substantially non-trivial.

So we will see. Personally, I think unless FB can start to show some really meaningful headway toward a 5B revenue in the near future, it's price will continue to go down. Also, given the negative QoQ revenue growth, FB may be in for a rocky road.

Password Security: When the Problem Exists Between Keyboard and Chair

Tl;dr When a third party asks for information that allows them to reset your passwords for other services, you should be very wary.

Technology Review has an interesting story about a potential security hole in gmail. Basically, it works like this:

1. Hacker site offers a user a discount/teaser/whatever and asks them to enter their gmail address.
2. Hacker site sends request to gmail to have an account verification code to the cellphone of the user.
3. Hacker site requests user to enter the verification code sent by gmail into their site.
4. Hacker uses verification code to hijack gmail account.

As the comments in that article point out, this is not actually a technical loophole. It is a prime example of PEBKAC: Problem Exists Between Keyboard And Chair. Specifically, one would hope that your average user would be smart enough not to give a password reset verification code to a third party that they had only just encountered.

Unfortunately, there is no foolproof way to protect gullible users from themselves, and it is far more important that the millions of gmail users who do properly use SMS password reset protocols have access to this functionality than it is to simply disable it for everyone. In other words, this is not really a security flaw, so much as an unfortunate externality of a very useful password reset system. In the same way that a bank safe cannot be secure if the bank manager writes the combination on the front door, no password security protocol can stand up to a user divulging sensitive information.

Moral of the story: anytime a third party requests secure information from another service you use, whether that is login info, password info, security questions, etc., alarm bells should go off. Before you supply any such information, try your hardest to determine the provenance of the requesting party.

Friday, May 18, 2012

Gmail, Why Do You Choose to Spite Me So?

I used to do a lot of front end development, so I find bad user controls to be particularly vexing. I doubt this makes me special, as it probably annoys most other humans, too. However, some things just pop off the page at me. The below is one of them.

Gmail, despite the fact that it is my lifeblood, is a frequent perpetrator of bad UI. This is often exacerbated by the fact that Gmail loves to push interface changes on its users without warning, explanation, and frequently without the ability to undo. I understand the necessity of this, but sometimes they get something infuriatingly wrong. Like this:
This is an image from my Gmail user pane (yes, everything of mine is space themed -- wanna fight about it?) where there used to be a button that allowed me to scroll to the second page of my Inbox. It looked rather like this:
In fact, I can get to those controls by clicking on the words "1-25 of many", which transports me to another Inbox view. However, from my default Inbox view itself, there is simply no way I can get to the second page of emails. I can think of reasons why someone might argue that this interface change is obvious, intuitive or helpful, but those reasons are all bad.

So, why in the name of Great Odin's Beard these controls are no longer in the default view is just flabbergasting to me. This may be a minor point, but it is still really sloppy user experience for a product that I spend hundreds of hours using, per year. If it is confusing to me (and it was), I can only imagine how less savvy internet users will react when this change percolates through to their inboxes.

Wednesday, May 16, 2012

Free Does Not Mean Valueless

Tl;dr People are unwilling to pay for baseline content; make your baseline content free and use it as a loss-leader to sell naturally scarce premium services crafted to engage your target audience.

A coworker forwarded me this article today, entitled "The Emperor has no Content."

It is a short read, and worth reading. Let me say that I am taking this opportunity, as prompted by Mr. Rutledge, to in fact, comment on my own site.

I disagree pretty emphatically with the conclusion that Rutledge seems to make, which is this:
If the content is valuable people should pay to read it.
This is a very common misconception. First off, it utterly conflates 'value' with 'money.' Many things that have value have non-monetary value. Second, it assumes that if something does not have a pricetag, it is worthless. This is also obviously incorrect.

More importantly, however, it misses the point: while I fully agree that advertisers probably do not have the interests of content producers at heart, and that online ads can be really intrusive and weaken online content user experience, I think that simply saying that you should charge for access is the exact wrong attitude. People are unwilling to pay for baseline content, and that is not going to change. What you can charge for is for additional services or goods on top of the free content, as a freemium model. People have no problem paying for things that are genuinely scarce, and online content is not scarce. If you charge me for access to news, I will simply read my news at a free source. There is a stronger argument for charging for access to creative material, as good stories are non-fungible. However, I sincerely cannot remember the last time I actually bought a book from an author who I had not previously read, and liked, and was given that original book by a friend.

It is interesting to note that intellectual property law actually gets this one right: You cannot own news items, nor can you own facts. That is outside the scope of copyright, as it should be. As a result, anyone who is arguing that newspapers should have paywalls is really headed down the wrong avenue: news is fungible. No matter what you are reporting, I can find out about it somewhere else for free. So charging for it just limits your audience.

However, if the NYTimes wanted to charge me for access to message boards that story authors actively contribute to, for the ability to ask questions directly of columnists, for the ability to upvote or downvote proposed story topics -- yes, I would pay for these things. Because they offer me some real value above and beyond merely charging me for something that has literally zero marginal cost to distribute.

Also, please note that Mr. Rutledge's car analogy totally, totally misses the point on this whole debate: Cars are a naturally scarce resource. Each one costs money to manufacture and uses physical material. Online news is not a scarce resource. Once you have written down a news story, it costs no more to send it to five people than to five hundred million. So this is the point that Mr. Rutledge has missed: rather than charging for access to something that is not naturally scarce, Mr. Rutledge should be focusing his energy on providing naturally scarce services or goods to his customers/clients which can then be charged for†. It's that simple. Mike Masnick, of Techdirt, has written literally dozens of articles on this exact topic. I'll link a few choicer ones here. Suffice it to say, I think that despite the fact Mr. Rutledge has very astutely come to some deep, and correct, conclusions about advertising, his underlying point is really misguided.

An Economic Explanation For Why DRM Cannot Open Up New Business Model Opportunities
Saying You Can't Compete With Free Is Saying You Can't Compete Period
In Which I Debate A Media Mogul Who Insists It's Crazy To Give Content Away For Free
Evidence Shows You Can, In Fact, 'Compete' With 'Free'
Economist: Copyright Is An Antiquated Relic That Has No Place In The Digital Age

So, basically, in conclusion, the attitude that things that do not cost money have no value is pretty vexing to me, not only because it is wrongheaded, but because it is so depressingly common amongst content producers, and it is an attitude that leads directly to shooting yourself in the foot. Because instead of figuring freemium services to offer that people would actually be willing to pay for, it is thumping on your pre-existing product and yelling at your consumers for not wanting to pay for it. So instead of being forward thinking and productive, one becomes interested in restricting access to pre-existing work. That, to me, is absurd.

† There are examples of people willing to pay high premiums for infinite goods. HBO is a prime example. So is a subscription to World of Warcraft. However, there is still natural scarcity in both these models: HBO's shows are really, really good and WoW is also really, really good (so I've heard). They are way better than the quality of many of their competitors, and I'm willing to pay for that. However, you will never get a similar disparity in news reporting. This doesn't mean that you cannot charge for premium access to elite content production of one form or another, but, again, it is really, really rare that you can successfully charge for pure content access. Additionally, WoW is not even really an infinite resource, because you are actively leasing server time and directly interacting with other people, who are themselves expending energy in their participation. Finally, ad-supported models can and do work. I'm perfectly willing to sit through Ads on Hulu to get my daily Colbert fix, and as Hulu grows, some of the ads are actually relevant to me.

Tuesday, May 15, 2012

Three Digit PE! / How much bigger can Facebook get?

Apparently, Facebook's IPO will put it in the fabled faery lands of companies with three digit P/E ratios. Specifically, it looks like Facebook may well have a P/E of 100 when it goes public. For reference, Apple's P/E is currently 13.6 and Google's is 18.55. If Apple's P/E was 100, it's market cap would roughly be 3.8 TRILLION dollars, and similarly, Google would be valued at 1.07T. Granted, P/Es that high are not unheard of, but, typically, they are very big when a company is still very early in its growth curve, or, on the way down, when earnings are particularly low but there is still fundamental underlying value to the company. For instance, EA currently has a P/E of ~66. This honestly is probably not great for EA, and, in fact, earnings are way, way down for the gaming industry as a whole right now, apparently to the tune of 25% March 2011 v. March 2012, and 32% April 2011 to April 2012.

So, no matter how you look at it, Facebook's valuation is really, really high. Given it currently has 900,000,000 users, and given that 80% of the world's 7B residents live on less than ~$3600 a year, it looks like the total addressable market for facebook is 1.4B... meaning that, globally, there are only around 500M potential users left. Accordingly Facebook can only realistically grow another 15.5%, if one just assumes linear extrapolation and ignores a whole lot of other details. I think that, in broad strokes, however, this rough estimate is actually quite telling. Facebook is so huge it already probably has the majority of its global addressable market. So unless it can start to figure out how to get a whole lot more money per user, they cannot rely on additional growth to justify their P/E, because they are approaching total market saturation. Update: Apparently, GM has just stopped buying paid ads on Facebook. According to the WSJ:
General Motors Co. plans to stop advertising on Facebook after the company's marketing executives determined their paid ads had little impact on consumers, people familiar with the matter said, a move that comes as more companies question the effectiveness of advertising on the social networking site.
I'm still betting on a first day bump for Facebook, but, hey, I may be wrong.

Tl;dr Facebook's valuation is 100x its yearly earnings, which is unsustainable given that they don't have much more room to grow, and that multiple is way, way higher than other tech giants.

Edit/update: A friend of mine pointed out that a mere $9 profit a year, per user, for Facebook, would generate 8.1B in yearly earnings. This would put their multiple at 12.34, which is actually pretty solid. So, that raises the following questions about ceilings:

1. If Facebook has 900M users, and is generating ~1B in year profits, is it possible to get to 8B in profits? To put it another way, is the way in which Facebook users interact with the site worth $9 per year, above costs, as opposed to the current $1 per year?
2. Of those 900M users, how many are actually active accounts? More specifically, is it already possible that Facebook has already reached a maxima of user-hours?

To reiterate, that's two questions: the first is about the value of user hours, the second is about whether Facebook can increase the number of user hours. Only time will tell.

Better Math on Online v. TV / Forgive Me, for I am Mortal

In an earlier post, I did a very rough comparison of the economics of television v. online users. I committed a pretty big mathematical blunder, and, also, found some better data. So let me correct myself.

Apparently, Pinterest users spend about 15 hours a year using the service, which would give a per-hour valuation of ~$1. This would value Mad Men at $42M, using the same math I did in the previous post. However, if you just calculated that Mad Men viewers have the same value as Pinterest users, instead of doing my elaborate (and probably useless) per-user-hour-valuation, that number is $57.5M

It is important to note that these are really from-the-hip estimates. In fact, the mistake I made in the previous post actually demonstrates quite well the problem with doing this sort of comparison: in order to get the value of 37¢ an hour, I just divided user value (16.50) by 50% of the hours spent on Facebook per year (48)†. The fallacy here is assuming that the value is unrelated to the amount of time users spend on the site; in other words, by this logic, if users were to spend more time on the sites, and the user value were held constant, the per hour value would drop. This is almost certainly wrong, as sites that users spend a lot of time on are inherently more valuable than those with a high bounce rate. In other words, the value of a site increases with time spent on it.

So, my point is that this is a really roughshod way of comparing apples to oranges, but it was the easiest way to do a back-of-the-envelope comparison between TV ratings and online users. I think my conclusion stands, however: it is amazing that a show like Mad Men, with 2,640,000-3,500,000 weekly viewers, would be a Network Television failure, in a world where having a few million users for an online service means you have a pretty big success on your hands. Many will say "sure, but internet companies are overvalued," and while I agree that this is probably true, I think the counterpoint is that the dated, legacy ways in which television shows make money undervalues them.

† The mistake I made was using the lower, not the upper, bound of estimated user hours to determine this value. I said I was being disfavorable in this calculation, which means I should have used the upper bound, not the lower one.

Tl;dr If online startups are overvalued, tv shows are probably undervalued, likely as a result of their aging business model.