Luis VillaRamblings on software, law, and the spaces in between.

Some friends have admitted to some confusion as to where I am living right now. More details in later posts, but the nutshell version:

1) Krissa and I no longer live in New York.

2) tomorrow night, we sleep in our fifth bed in seven nights.

It does get a little more normal from here on out, sort of. Tomorrow night’s bed is at Krissa’s dad’s place in California, which we will call home for the rest of September while we prepare for the wedding and honeymoon. October and November we will be out of the country on honeymoon. Best bet for contacting us during this whole period is direct email; twitter, facebook, phone, etc., just aren’t going to cut it, and messages which are sent over mailing lists or other email which does not have my email address in the ‘to’ or ‘cc’ field is likely never going to be read.

In preparation for a honeymoon in places where data access is expensive I’ve put all of wikitravel on my Android phone via the OxygenGuide project. I’m not sure how useful it will prove in practice (especially since all maps are stripped out and there is no search) but this seems to me to be an obvious next step for travel guidebooks- everything in your pocket, searchable (at first by text and later by GPS coordinates), etc. It certainly is not a straightforward process right now, and eventually it will be obsoleted by cheap, global data access, but in the meantime I look forward to using it on this trip and in the future.

In the meantime, I’ve also become a bit intrigued by other custom uses. I was just at the wedding of a friend in an exotic-ish location; as part of the gift bag that many wedding guests put in the room of their guests there was a very nicely done printed version of content from the city’s wikitravel page (which also included bios of the bride and groom, a schedule of events, etc.) I may do something similar, albeit probably just by emailing pdfs to people :) Will be interesting to see how this and open street map integrate in the future.

I have to admit (and I’m curious to know if others have had this experience) that I’m reluctant to rely on the hotel recommendations in wikitravel; not sure why that is- probably just that it seems fairly unsystematic and prone to manipulation?

Is there a state of the art for free software project bounties? I’m sort of curious, because I’ve become a heavy user of a project which has an overworked maintainer and no particularly vibrant community.  I also  have no time/ability/desire to dive into that codebase, but I have two features that I’m pretty sure upstream would accept if someone coded them up. So I’m sort of curious about the options for situations like that, but realize I haven’t looked at the problem in ages and don’t know what the state of the art is, or if any one is even experimenting with it anymore. Anyone? Bueller?

I like the concept of nerd merit badges- I even want the inbox zero one- but the execution is a little weak. Seeing this started me pondering on what I’d do if I were going to make my own.

The temptation, of course, is just to have a logo per organization- gnu, gnome, moz, fedora, etc. But the core of the merit badge idea is that you’re indicating skills learned, not organizations. (Yes, I’m an old boy scout on top of all my other nerdiness.) So I tried to think along those lines, and here are some of the skills I’d like to decorate my bag/laptop with:

• GTD: a david allen head, natch.
• emacs: emacs logo?
• bugmaster: old gnome bugsquad logo? (mosquito with crosshairs)
• infolaw: scales, or something (yeah, I know… pretty weak.)
• perl: a very short obfuscated perl program
• peer production: (because open source just isn’t right for some reason) (but hell if I know what that logo would be; nerdmeritbadge uses github’s octocat but that seems… weak)
• lego: you know it!
• family tech support: nerdmeritbadges has this one, and it isn’t a bad concept, but their logo for it is terrible.

I bet you could do a passable job of them with this cafepress option and some decent gimpery, but if people have decent suggestions I’m all ears. :)

As I mentioned in my lwn interview a few weeks ago, I’m curious about where the Open Collaboration Services/Social Desktop is going- while I have not been able to figure out if this is the right way to do it, it is obvious that the Free desktop needs to start experimenting with and exploring this sort of space, and this project seems to be leading in that direction.

That curiosity has now turned into a bit more; I’m going to be a judge for the application contest that Social Desktop is now running. The contest closes in late August; could be a fun way to spend part of your summer hacking time.

While I’m obviously not going to be hacking, I expect this will be an interesting educational opportunity for me- I want to learn more about what these APIs can do, and this should be a good way for me to do it. :) More importantly, this is an interesting way to get eyeballs and hackers focused on this space, and I hope it (or something like it) succeeds.

Early this week I published Columbia Science and Technology Law Review Volume 10, completing my responsibilities as Editor-in-Chief.

We have some interesting articles that I’m glad to have worked on – particularly, I think, the very interesting (and readable!) article on what is known in contract law as the ‘mailbox rule’- and how it is obsolete in the internet age.

Of perhaps more personal pride to me is the letter from the EIC that I was able to write, explaining that the journal is going Open Access and complying with the Durham Statement on Open Access, recently published by the librarians of several leading law schools. In practical terms, this was not a huge change for us- we already published online and in pdf. However, moving to library servers is important for our permanence, and it is an important symbolic change as well. Frankly, I expect that the economic hit many law schools are taking right now, combined with new accreditation standards that are likely to reduce or eliminate mandatory journal purchases by libraries, is going to push a lot of journals away from expensive paper publication and towards online publication, and my hope is that we’re ahead of the trend here- showing others that you can still do good scholarship this way.

We’ve got a variety of news coverage out of it, most notably from Open Access News and Berkman.

I can’t say I’m completely thrilled with the process; specifically, we decided to go with the more conservative non-commercial no-derivatives Creative Commons license, when I would have preferred a more open share-alike policy. But we’re talking baby steps here, and I think even this restrictive license is good for our authors (open access increases citation counts) and good for law as a whole.

And with that I’m done with STLR, and really with Columbia. Thanks to my entire staff; you were great under all circumstances (pleasant and otherwise) and to Krissa for putting up with me for another year. :)

Usually, I’m above saying ‘I told you so’, but sometimes… well, lets just say that it is clear that most of the people commenting in this bug about the ‘confusion’ caused by the Ubuntu One mark didn’t read my old posts on trusting open source companies or my (old, not terribly good) trademark paper. To paraphrase the paper:

Trademark licensing centralizes control with the mark “owner” instead of decentralizing control with direct contributors- i.e., the copyright holders. Th[is breaks] the community’s implicit social contract, formalized in the source’s copyright license, that control is to be decentralized…

As the paper describes in more depth, unfortunately, unlike copyright, where the default is that you have to do an assignment to lose control of your contribution, in trademark, the default is that you never have control- some central ‘source’ body does. And as a result, you get situations like this one.

Moral of the story, short version: don’t get too attached to the purity and value of a mark which neither you nor representatives you elect have control of.

[Added a few minutes later: I should point out, by the way, that Ubuntu probably has the best trademark policy of any open source project that I'm aware of- it is far and away the most respectful of fair use and non-commercial use. But at the end of the day, because the mark is owned by a body which isn't legally controlled by the community, things like this can happen, and probably should be expected to happen.]

The world needs more lawyers like this:

I often explain to businesses that the main reward for a great, original product is a succesful business based on that product, and that intellectual property notwithstanding, the best way to protect most great ideas is by consistently excellent execution, high quality, responsive customer service, continued innovation and overall staying ahead of the competition by delivering more value.

Fender, according to the record in this opinion, understood this well for decades… Only in the feverish IP-crazy atmosphere of our current century did the company deem it “necessary” to spend a fortune that could have been used on product development, marketing or any darned thing on a quixotic quest for a trademark it never believed in itself.  That is more than in impossible dream — it’s a crying shame.

I remember telling a room full of execs in 2002 (in the context of a discussion of Lego’s threatened use of trademark to restrict legOS) that the correct response to criticism of their product of the internet was to build better products. They looked at me like I was a naive child. And I suppose it was in part naive; there are going to be crazy people who criticize no matter what, which I’m not sure I understood at the time. But overall I’m glad to see that others (even lawyers!) still think I’m not completely nuts.

(See also Matt Haughey recently on a basically related topic.)

Some anecdotes on data loss in the cloud:

• online backup provider loses data, sues systems provider
• Nokia’s Ovi service loses user contacts and images
• ma.gnolia lost all user data

A more complex case, where the provider did (mostly?) the right thing:

• Yahoo closing its locker service after 10 years

These are not necessarily straightforward; each failure has disparate causes (too much reliance on expensive hardware in the first case; too little, perhaps, in the ma.gnolia case). But understanding what happens and when here is important as more and more of our data moves cloud-ward.