New DivaBlog

Ada was a Lady

Augusta Ada Lovelace was arguably the first tech writer, and almost certainly the first computer geek groupie. Some folks arranged an international Ada Lovelace Day Blogfest today to call attention to Women in Technology, and I agreed (some time ago) to participate.  Probably I would have spaced it, but a handy nag just came up in Facebook (...see? FB is good for something besides re-finding your 7th grade boyfriend afterall...).

Okay, here's a blurb about a woman in technology who I very much admire:

This is a picture of Allison Randal (taken by the wonderful James Duncan Davidson...the original is hosted here on Flickr). 

I think Allison is the bomb...seriously.  She's served as the President of the Perl Foundation.  She's long been one of the program chairs for OSCON.  She's a code contributor to CPAN, the Parrot project and I suspect some other projects as well...*and* she still manages to be quite beautiful and poised through all her geekiness.  She's fun to hang out with (although now that she lives in South Africa I'm not seeing her so much around the playground).  Have to say I admire that whole "I'm moving to South Africa" thing too...I remember being that brave at some point in my life.

So, there you have it.  My Ada-Day heroine is Allison.  Who is yours?

Lunch meeting with Alolita Sharma

Having lunch with my pal Alolita, di
Discussing opensource.org content overhaul. Nice to have her energy on our side.

Pre-OSCON Frivolity

Monty and Simon planning World Domination

Google Android first real "customer" for Apache Harmony

I was gonna blog about this topic from scratch, but this post by my Apache colleague, Stefano Mazzocchi, does a good job of explaining how Android makes use of the Apache Harmony class libraries with a new, Open Source VM called Dalvik.

Given Sun's continuing refusal to actually address Geir Magnussen's Open Letter concerning TCK licensing with anything other than marketeering, this is both a brave and perhaps a brash thing for Google to have done...and I for one can't wait to see how or if Sun responds.

You might think their response will be swift, but sometimes Sun has a "Bambi in the Headlights" reaction to game changing moves in the marketplace (witness the many months that lapsed between IBM's first announcement of Eclipse in November 2001 and Sun's eventual response).

Open Data @ FOO

There are no less than four sessions about the concept of "Open Data" at this year's FOO (last year there was one). 

Open Data is something I've been interested in for a couple of years now as a natural extension of Open Source (not unlike Open Standards).  The idea is that your online data should be available for your use freely.  There are still many large datastores that treat data you submit (or that is gathered about you) as a "proprietary resource".  So for instance, how easy is it to manage your photos online?  If you get tired of one photo aggregation system, how easy is it to transfer your assets to another system?  Do you "own" those photos?  You certainly own the copyright, but do you own them from a practical point of view if you can't manage them? 

I'll make more comments after I've attended the sessions.  I'm especially interested in the session led by Mark Hedlund.  Also, Julian Cash is here and will no doubt be talking about Move My Data.org.

Stalking Tantek Çelik

I must here take a pause to admit a private vice.  I am stalking Tantek Çelik.

Every day my cellphone beeps new SMS's that tell me Tantek's whereabouts.  Its gotten to the point that my husband is suspicious of my interest in Tantek.  I mean afterall, he's one of the Top Ten Male Hotties Of The Web.  But that's not how I came to be his stalker.  I run into Tantek at conferences, but he's nearly always surrounded by people.  I wanted a more singular conversation.

So, last year when Dodgeball started to get popular in San Francisco (before the Google acquisition that seems to have gone awry)...I signed up and asked Tantek to be my contact because I was hoping to figure out a less congested place to meet up with him.  I'm big on tea dates at Samovar.

What I wasn't counting on was the sheer enthusiasm with with Tantek broadcasts his daily movements.  It's not uncommon for me to get 10 messages about Tantek in a 24 hour period.  I don't know that much about the whereabouts of my kid.

And I've gotten to where I worry a bit about Tantek (as a Mom...heck, I'm nearly old enough to be his Mom).  So for instance, I'm happy when I see that he's at Mission Cliffs or munching brown rice sushi at Whole Foods.  But I think he drinks too much coffee.  And I worry when he's on the occasional pub crawl...will he make it home in one piece?  On the other hand, I've learned about a few places from him, and I've found that we go to some of the same places.

The really funny thing is, I've still never met with him.  Tantek, if you're reading this...can we just make an appointment?

Linux Foundation Collaboration Summit (day one- afternoon)

After the usual yummy Google Cafeteria lunch...on to the afternoon panels.

Can I first offer a couple of hopefully helpful (if unsolicited) public comments to the organizers of this event?  Chris DiBona recently blogged his frustration with participating on panels, and I have to say I agree with him.  People like us are asked a hundred times a year to be on panels, and there are definitely better and worse ones.  Moderator-driven panels are not nearly as interesting as audience driven panels, but conference organizers keep forgetting this.  The energy-level in this room goes up when the audience is allowed to ask the questions and down when the panelists act like "talking heads".  Also, cutting general mingle time down in favor of more panel time is a mistake.  The interesting thing is the mingling, not the lecture.  One reason I love O'Reilly conferences is that they have comprehended these facts.  So, in future please allow more mingle time in the schedule (night-time "fun" events are not mingle time...nobody wants to talk about work at them).

The first afternoon panel was about getting more applications onto Linux.  Lots of Industry folks (see schedule).  There were comments about the usefulness of the Linux Standard Base (they all liked it...but afterall they are on an invited panel ;-) ).  There were snipes about still having to "pick a target" when app vendors choose which versions of Linux to validate and support.  Lots of questions about how to make money.  Quote of the panel was Mike Milinkovich Executive Director of Eclipse.org saying..."If you're a small ISV and your only platform is Linux, it SUCKS being you"...(he did have the good grace to say that he was hoping to get out of the room alive after the comment).

Next panel was the obligatory legal one...this one about GPLv3 and its effect on the Linux ecosystem.  Interests (in to GPLv3) included patents, standardization, fragmentation and aggregation such as Linux Foundation.  IMHO, there was more than a little pandering to secular interests (adverts for initiatives championed by their employers).  I personally asked a question about Open Standards and the need for a definition there, which Andy Updegrove agreed with (no surprise, but it was nice).

At 3:00pm, I had to take a call (it's a miracle my cellphone didn't call me away before now)....

Linux Foundation Collaboration Summit (day one- morning)

Sitting in the audience for Day One of the June 2007 Linux Foundation Collaboration Summit (they should number these things). The room is full of Linux developers, open source luminaries and corporate interests trying to have conversations about what would make Linux "better". (Here is a pointer to the published schedule, for quick reference).

Jim Zemlin started off with a few choice "adverts" about Linux that showed how the overall FOSS community is seen by the outside world, including: this one and this one.

Next was a panel with Kernel Developers. Within that panel, there were questions about provisioning device drivers (in which Intel's PowerTOP was applauded and Intel was generally given high marks for delivering drivers to Linux (yeah!)). Also discussed were the topics of accessibility, non-english participation, and Linux for mobile devices. Despite Jim Zemlin's pleas not to "make sausage" in today's sessions (I wish he would stop saying that)...there was a bit of heckling (just showing again the friendly and informal nature of the overall FOSS community we love so much).

Second panel was representatives of Industry interest in Linux. Dan Frye said that IBM is really interested in Real Time Linux and that they like to think of themselves as members of the Linux Community. Christy Wyatt from Motorola said that although Moto is a somewhat fusty company [my words, although implied by her], they are working on Linux for Mobile Devices, especially in the area of tooling through Eclipse Foundation (which she invoked more than once). My friend Chris DiBona said that Google is dependent on Linux and tries to give back through projects like Summer of Code and hosting events like the one at which we are met. Lengthy discussion ensued about how good things are at IBM, Motorola and Google. Christy Wyatt especially had a lot to say about the importance of "owning your own ecosystem" (although she tried later to respin those words as "enabling"...oops ;-) )...shows that she wasn't lying when she said Moto has a lot to learn.

I predict that Dan Frye is going to get quoted a LOT (and the FSF should make a tee-shirt ) about his great comment when asked for the top two things the community can do for Linux..."When the GPLv3 is final...just CHILL !!!"

Last before lunch was the always sublime Mark Shuttleworth. Much of the content of his talk has been exposed in detail in his series of blogs about the big challenges in the FOSS Community. He talked about the importance of "Collaboration", including aspects such as managing "poisonous people" who monopolize conversation and effective working between projects. In general, he said he thinks that Federation and increasing Decentralization are the answer and called for new tools to be designed with these memes in mind. In terms of things that need to change, he said that translations and bugs and patches aren't "moving upstream" (I think I'd call this "flowing" because the prevalent "upstream" image implies a hierarchy and there's already so much of that). He called for better conduits that can broadcast completed work around the whole ecosystem. He also mentioned issues of authentication, and asked how we can remove other barriers to collaboration. Since he is a fan of "Federation" he pointed out that Standardization probably isn't the answer (is LaunchPad broken by design?), but he believes that better Project Management will be essential in future. Projects that are doing a good job? Mark thinks that GNOME is very well self-organized and does a good job of communiciating with other projects. For the Ubuntu project, Mark also took Mozilla as a design center. He pointed out that Microsoft's internal systems allow their internal developers to find and have conversations with peers very easily, and that similar inter-project equal participation is lacking from the FOSS world. As always, he pointed out lastly that we should all strive to focus on that on which we agree.

Okay, okay...now time for lunch (more afterwards)

OSI Board considers Attribution Licenses

Many companies over the last couple of years have started "Web2.0" businesses (I'm using the celebrated term to mean they are offering metered services via web-applications which they have authored and which they host).  Some Web2.0 Application companies want to also call themselves "Open Source" although their source code is available only under *modified* copies of licenses that are "OSI Approved".  These modified licenses are most often derivatives of the Mozilla Public License.  A partial list of the most visible of these include SugarCRM, Compiere, Alfresco, Socialtext...to name a few.

There are several potential issues with these licenses.  First and perhaps most troubling of all, modification of licenses is supposed to trigger a re-submit to OSI before claiming that the resulting derivative is still Open Source.  OSI has long had a policy of only considering licenses which have been submitted by their authors and so we have been working for some time with several of these companies to get them to publicly submit something we can deliberate upon.  Meanwhile, some members of the community are getting restless.  There have been queries by community members who are concerned, and today Nicholas Goodman submitted a "template" of an Attribution License to License-Discuss to try to get the issue opened.  We believe however that we are on the verge of getting one of the companies in question to submit their actual license, which we would prefer.

Here's the thing: We have no reason to think that any these companies is intentionally acting badly.  It is not in anyone's best interest to punish a given company for being willing to submit their license, and all have expressed concern that the License Discuss discourse would be harsh and perhaps unjust.  They have all read the OSD very carefully and all believe they have made modifications to the MPL that shouldn't violate the rules.  They all point out that the exigencies of the application space are different than with infrastructure software such as the Linux kernel.  Also they note that we have already approved licenses requiring some form of code attribution (most often a permanent notice somewhere in the sources).

Many of these new Web2.0 company Attribution Licenses go a little further however and require some sort of attribution in the form of a logo or text which must be displayed on web pages built using any part of the covered code.  In our deliberations we are trying to determine how much attribution is "too much" to inhibit the Open Source Effect fueled by unintended consequences and maximum code reuse and where to draw a line if indeed a line needs to be drawn.

We'll continue to discuss these issues and to work for submission of a license to License-Discuss for public comment.  But we want to use the process constructively, not punatively.  We always learn something when a new license is submitted.  Our process is about learning and examining the Open Source Effect as expressed in licenses.

Java Attribution Timeline Edits

Thanks to Eduardo Peligri-Llopart, Mark Wielaard and Ed Burns who added milestones to the timeline.  The original post from earlier today has been updated to reflect all comments received so far, and I got a suggestion to move the timeline to java.net, which I am considering.

Geir and I had a very nice time at Sun this morning listening live to Jonathan Schwartz and Rich Green's comments (which were also webcast) and then hanging out on IRC and later on Second Life with the Open Java team.  It's a deja vu experience to be at Sun where we held the first Open Source Summit in 2004 a few months before I left Sun.  Many of the folks hanging out to cheer for Open Java today were not big fans back in 1999 when we started talking about the idea; the room was certainly all smiles today.  My, how time heals all wounds :-).