Archive for September, 2007

GPL Heads To Court In US

Yesterday the Software Freedom Law Center (SFLC) announced that they will be acting on behalf of the authors of Busybox in a copyright violation suit against Monsoon Multimedia a manufacturer of electronic devices. The filing (in a pdf) is here. The SFLC, if you haven’t heard of them before, is a legal organisation which offers support to the FLOSS community. Eben Moglen, co-author of the GNU General Public License, is its Chairman. Busybox is a free software project released under the GPL that provides miniature versions of common UNIX utilities, especially useful for developers of device-embedded software.

So what’s the problem? Well, its first public stirrings came in a thread on the official discussion forum devoted to Monsoon’s video streaming device the Hava. The thread-starter asked if the device ran Linux. Other posters soon chipped in saying that, as far as they could see, it ran a system based on Busybox and the Linux kernel. This was somewhat surprising, as the Hava web site does not mention this fact, and no source code to this software is provided, as one would expect if Monsoon were distributing GPL’d software in compliance with the licence. Soon a moderator on the board acknowledged that there was indeed GPL’d software in the Hava, and that they would be trying to organise source distribution as soon as possible. He added, however, that the posters who had discovered the GPL’d software had done so in violation of their End User Licence Agreement…

You can read the entire exchange here. If the posters (including the Monsoon company moderator) are correct and the Hava contains GPL’d code, then it is difficult to see how the SFLC’s suit can fail. In fact, it would seem to be be such a cut and dried case that it makes me wonder how it ever got as far as the SFLC filing suit. The usual outcome in cases such as this is that the SFLC (or previously the FSF) would inform violators of the GPL of the problem on behalf of the software’s authors. The distributor would then - almost without exception - remedy the problem. In most cases this would simply involve making source code available from their web site. The process worked so well that it was almost a problem in itself; when people ask whether the GPL is enforceable in court there is an almost complete lack of precedent.

Some opponents of the GPL argue that it is a contract, and that further it is a potentially ineffective one as there is no overt agreement on the part of those who take up GPL’d software. Proponents often point out in response that GPL’d software is a copyright work, and that the default situation is therefore that you cannot adapt, copy or distribute it without a licence. The GPL, they say, is a possible licence, provided to anyone at all who chooses to conform to its conditions. You can take it or leave it, but if you leave it you have no route to legal adaptation or distribution of the software. Nevertheless opponents continue to argue that the quid pro quo of ‘licence for conformance’ does constitute a contract…

In practice, of course, all of this argument is profitless. The fact is that up to now people have behaved as though the GPL is an effective tool for ensuring software freedom, and to that extent it is, regardless of the near-philosophic discussions that continue to spiral around it concerning status, consideration, performance and rescission. The case against Monsoon is unusual in that it is a real public test for the GPL and the widely held view that those who violate it can be legally pursued using only copyright law. Perhaps the SFLC felt that it was time to develop some real precedent on the issue. Whatever the case, both sides of the debate will watch this case with interest.

SHOCK: A social networking tool I like

On a number of occasions I have blogged and presented my concerns about current social networking sites. Some of my main concerns revolve around the fact that my email client is my main online communication tool and I don’t want to work with a suboptimal communication channel (although I’m happy to augment eMail where appropriate). Furthermore, I don’t want to maintain my network manually, in multiple places and in an externally owned data store. The vast majority of this information can be derived from my mail communications logs, networking sites shouldn’t need me to manually recode this data.

Despite these concerns, I do believe there is some value in the emerging social networking environment. I’ve been returning to take a look periodically, since creating my first account on Orkut in 2002. Until recently I’d not seen anything to blow me away.

LinkedIn is a site aimed at professional networking. It’s been around for quite a while, and has grown steadily in terms of users, but not in terms of silly gadgets. Of all the social networking tools I find it the most promising, although still deeply flawed.

It is minimalist and relies on human to human communications to build the real network. People who believe a valuable network is one in which “friends” are people who “watch” you do things, as opposed to actively participate with you, or people who have the time to waste on frivolous toys like “poke” will not like LinkedIn. However, people who feel that a valuable network is one with expertise that can be drawn upon when appropriate will find it useful.

Despite the narrow focus of LinkedIn I was still frustrated with the need to manually maintain my network. I simply don’t have the time to search out everyone I make contact with and add them to my network. However, earlier this week I spotted the LinkedIn EMail toolbar. Unfortunately it is only available for MS Outlook, but I decided to set one o f my accounts up in Outlook and give it a go.

I was underwhelmed by Outlook (in around 7 days of use it has crashed twice), but blown away by the features of the LinkedIn toolbar…

When I get an email from someone I can instantly see if they have a linked in account, complete with a summary of how they are connected to me. This happens right where I do my communications, in my email client. If the author doesn’t have an account I can invite them to my network with a simple click.

The toolbar also scans my mail periodically and, via a dashboard page in my email client, tells me of any new contacts that I can add to my network (and local address book of course). There are other useful features, such as having it nag me to communicate with a contact within a given amount of time, the nagging stops once an email is sent.

This is what I call networking. People interacting with people as individuals rather than the faceless masses. I deliberately avoid the word “social” here, LinkedIn is more about business networking than social networking. Now, if only a social networking site would learn from this approach.

Perhaps the Mozilla Foundation, who recently announced they are to put effort into innovation in the email client will help leverage eMail in a web2.0 world.

ICANN to start Internationalised Domain Name testing

The ICANN plan to roll out domain names non-western scripts is about to release trial top-level domain names using the world ‘test’ translated into Arabic, Persian, Chinese (simplified), Chinese (traditional), Russian, Hindi, Greek, Korean, Yiddish, Japanese and Tamil. The top level trial domains will be retired once production domains in these scripts are rolled out:

It is planned that the .test labels will be kept in the DNS root zone and resolving with example
positioned at the second level (i.e., translations of example.test) until registrations in a
corresponding script are available in a production environment. Although it is anticipated that
the evaluation facility will be of short-term utility the lifespan of the evaluation may be
extended if it is demonstrated that target groups will derive continuing benefit from it.

I’m expecting that many, many, applications which touch DNS are going to have to be patch to fix bugs that will be shown up by this change, so expect a wave of patches and updates 2-6 months from now. Well behaved applications which handle DNS by calling system libraries are likely to be OK, since this change have been in the wind for ~3 years and the maintainers of specialist libraries should be prepared for it.

I’ve previously written about ICANN’s failure to move on this and other topics, but it looks like I may have been too quick to criticise them.

Users Report and Feedback on the JISC Blogs Pilot

At the recent JISC Services Communications Group Upskilling Day, I gave a presentation on our use of this, the JISC Blogs Pilot.

My slides have been up on the Comms Group wiki for a while, but that’s password protected, so I’ve published them on the OSS Watch website too, as PDF and HTML.

In summary, the most important key to having a successful blog is knowing who you’re writing for, why you’re writing for them and what you’re writing about. The pilot service had a number of short-comings which Matt Dukes assured us in his presentation are about to be rectified when the service goes live.

Email management: Saving searches in Thunderbird

At the JISC Comms upskilling day yesterday, Ross described how to create a saved search searching for occurrences of your name across all mail folders in a mail server. This means that all messages from mailing lists can be filed into per-list mail boxes using filters and then those mentioning your name re-aggregated into a virtual folder, so you don’t miss emails that are follow up to something you posted weeks ago or emails to obscure mailling lists when someone mentions me by name.

Go to File -> New -> Saved Search … , fill in the dialog box and choose which mail boxes you want to be searched.

The saved search dialogue box

Despite the warning, I’ve not found “Search Online” slow (but then I only have a few thousands of emails on this server).

Saved searches in thunderbird do not copy or move the emails as filters do, so when you read, delete or flag the emails in the virtual folder that the saved search creates, you are actually acting on the emails in their original folder.

Your dialog may look slightly different to the one pictured here, which is from Thunderbird 1.5.0.13 on Ubuntu.