Older blog entries for pjcabrera (starting at number 17)

Let Java Go! Petition

As some of you may know by now, Eric Raymond has posted an open letter to Scott McNealy, CEO of Sun Microsystems, pleading with Sun to open source the Java platform. A colleague of mine and I want to do something about this.

We are eXtreme Programmers to the core. Using his customer stories describing what he wants to do, I have started writing a simple J2EE app, using the Struts and Hibernate open source Java frameworks. At its most basic feature set, the user simply enters their email address and name on a web form, and after the user clicks submit, the app emails a copy of Eric Raymond's letter to Scott McNealy, James Gosling, and the Sun Microsystems Board of Directors.

We want to let Sun Microsystems, and Scott McNealy in particular, know in a very personal way, how many developers want Java to be open source.

We need your help. We do not have at the moment, reliable hosting that can take the possible beating of a Slashdotting. My simple Java app may not survive the onslaught on my personal home server. :-)

If anyone can provide Tomcat and MySQL hosting ASAP, I am willing to Paypal some money each month between now and the last day of OSCON 2004 in Portland. I want to hit Sun with a thousand emails of Eric's letter every day, between today and OSCON 2004.

If you can help, contact me at:

pjcabrera AT sociedadjava DOT org

Hey, Scott!

Let Java Go! Let Java Go!
Let Java Go! Let Java Go!
Let Java Go! Let Java Go!
Let Java Go! Let Java Go!
Let Java Go! Let Java Go!
Java Society of Puerto Rico

I have been involved with a volunteer-run Java Users Group in Puerto Rico for little more than a year. During that time, I have been giving free monthly workshops and presentations on applying Java open source technologies. The contents of my presentations is distributed under a Creative Commons Attribution-ShareAlike license.

The JUG has finally organized a board of directors and an executive commitee to carry out activities. I am vice-president of the board, and executive in charge of recruitment.

As part of a membership drive strategy with the local four-year colleges and universities, we plan to hold RoboCode and Java TetriNet competitions in May of 2004.

For more details, see my JRoller weblog.

The source code for competition enrollment, online score posting, and tournament-like eliminations will be distributed under an Apache license. [ I had to work open source in somehow. :-) ]
16 Feb 2004 (updated 16 Feb 2004 at 20:44 UTC) »

I absolutely agree with you about large hardware store chains and the little local mom-and-pop varieties. I bought my PowerBook from a local Puerto Rico Apple shop that has been struggling for over ten years, instead of buying from CompUSA or over the web. I believe in choice, and if I don't support the little guys, the big guys will soon be telling us how it is. Fuck that! :-)

I also find myself in agreement with you about people from other countries working in a country and not knowing the language. But a little perspective is needed, please.

Some of these folks are escaping oppressive regimes and they come to make a fair living, free of hate and persecution. Who are you to give them any lip? They are probably having a hard time working full-time and learning the language on their own time. How would you like to be bitched about when the only place for you to run to doesn't speak your language?

Besides, you will soon learn your own employer is either giving them your job overseas, or bringing them over to the USA to take your job off your own hands. I'll love to see your perspective on this in 5 years, when the baby boomers retire and the USA finds itself short 14 million professional-level employees.

One question, deek? How many languages do you speak? Could you live in Japan if suddenly Communists took over the Congress and the White House and the only place to find work was Japan?

That's what I thought, bitch.

By the way, you just got bitched at by a multi-lingual Puertorican (6 languages and counting.)
16 Feb 2004 (updated 17 Feb 2004 at 00:50 UTC) »

repeat send of another post. removed

16 Feb 2004 (updated 17 Feb 2004 at 00:51 UTC) »

repeat send of another post. removed

Jiggety-jig, jiggety-jig, back again, back again.

I'm not going to make excuses as to why I haven't updated my blog in the last seven months. No one really wants to know anyhow. :-)

I'm back with a new focus on Java free software. I will be working with Java free software on the job and in other professional level projects, so there will be a lot more free software news coming from me now. I'll write about that at some other time.

Now moving right along...

The pond just got a whole lot bigger for the little folks...

I find Red Hat's leaving the consumer operating system market to concentrate on the enterprise market awesome news for the growth of Linux, but it's a bit sad. It marks the end of an era and the start of a new one for Linux. The most famous Linux distro is leaving its roots behind, maybe not in spirit, but at least in writing. :-)

The interesting thing is, this now leaves room in the consumer market for Mandrake to come back from bankruptcy. Now that the big fish has moved to other waters, suddenly there is a huge market gap to fill.

Of course, Mandrake now has even more competition than before, with Lindows and Xandros also positioning themselves in the consumer Linux market. Lindows goes for the computer newbie, while Xandros goes after the early adopter hardcore gamer that doesn't want to leave Windows games behind, doesn't know how to hack Linux either, they just want to move to this Linux thing because it's the cool new thing. I'm not sure which one has the higher market, but as Mandrake is the closer one to what Red Hat was, it may be the one to gain the top spot. I think Lindows and Xandros are positioning themselves in really narrow niche markets.

As for Fedora, I hope it turns out regular releases that compare in stability to the "old" RH Linux. I also hope it stays a volunteer-run project and doesn't go commercial. The world needs a showcase volunteer-run distro that doesn't require one to be a hard-core Linux hacker.

Disclaimer: I myself use Gentoo and Linux-From-Scratch for my development servers, so I'm not bashing hard-core distros. All I'm saying is that there are a lot more mainstream Linux users than hard-core Linux users, and I'd like Fedora to be there for the mainstream Linux users the way Red Hat's consumer Linux used to.

My best wishes to Red Hat and Fedora in the years ahead.

A month on the lam...

A little over a month has gone by since my last post. I got swamped with work from my job, and barely had any free time for hacking. Things have been back to normal for a week now.

Here's the summary:

Free Software in Puerto Rico Government

During this last month, the March 12-13 conference promoting broader free software use in the Government of Puerto Rico, was scaled way down in scope, because of lack of financial support. Now that this has passed into history, I am starting to plan a 'real' conference for some undetermined time in the future.

But first, I am exploring commercial sponsorship possibilities from the green-lizard and rouge-thing-on-the-head free-software companies, as well as the two-lettered computer hardware company and the three-lettered computer hardware company.

Mono and Gnome Hacking

I cleaned up some documentation in The Mono Handbook and began adding content to the introductions of the various ECMA standard .NET namespaces.

I am having a blast rewriting Same Gnome in Gtk# and Glade2. I didn't know Gnome2 programming could be so easy! :-) I am documenting my effort, to be included in The Mono Handbook as a case study of Gnome programming in Mono.

After I'm finished with Same Gnome, I'm converting GnomeCalc. :-p

Hacking the Linux kernel

My only solace during this last busy month was reading Understanding the Linux Kernel, Second Edition, during my lunch break and before bedtime. I also located a copy in good condition of Linux Core Kernel Commentary (aka LCKC) with the CD-ROM, which is now out of print. Finding this last book in good condition was a waste of time, because after only a week of use, the spine broke and the pages started coming unbound, which seems to be the fate of all LCKCs in existance. :-/

Good hardware is so cheap these days, I went and built a new spare PC, installed Debian GNU/Linux on it, plus the source of the last three releases of the 2.4 series kernel, some of Con Kolivas' patches and his contest benchmark. I grabbed the last three releases from the 2.5 series kernel as well, to keep track of the latest development. I also have installed LXR cross-references to all this source on my XBox Apache server.

Since I don't have the time to read the lkml every day, KernelTrap and Kernel Traffic will be my new found friends.

It seems in my overenthusiasm, I misunderstood my invitation to the free software in government conference. I was asked to 'participate', and I assumed that meant that I was asked to speak. More details later.

Talk loud, and carry a big stick ...

I have been promoting free software a lot in and outside of work in the past two years, and it has apparently attracted the attention of someone at the Office of Management and Budget Planning of the Government of Puerto Rico, the country where I live.

I've been asked to speak at a small conference discussing the greater use of free software in the local goverment. In typical fashion, I have only a month to prepare. I received the invitation yesterday, and the conference is scheduled for March 12 - 13.

I have never given a talk about free software to a group bigger than 6 to 8 people, and even then, I was preaching to a mostly devout choir, so to speak. The idea of standing in front of government paper pushers and showing them how free software is good for their budgets is scary (hence the title of this entry.) But the argument is an easy one to make, so I am not worried especifically about what I will say. I am worried about my delivery and effectiveness when speaking to a group so unlike myself.

With only a month to prepare, it's going to be difficult for me to drum up support from important companies in free software. I do not mean direct financial support to myself or to the conference, but support as in them bringing demo units showcasing their software, especially its Windows and Office compatibility.

Perhaps this conference should only be seen as a stepping stone, and I shouldn't worry so much about making a big impact so early in the game. After all, it seems the conference planners are mostly convinced about the financial benefits of free software, and only want to bring this information to the attention of their colleages.

I am meeting with the conference planners this week and next week as well. Perhaps we can start planning a bigger, better island-wide conference in the summer or fall. We'll see what develops.

For the next week or two, I will probably scour the web for summaries or transcriptions of similar talks and conference proceedings in the last year, to aid me in preparing my presentation. In typical free software enthusiast fashion, I plan to reuse other people's work and strengthen the stew with my own views.

If anybody has any pointers, suggestions, comments, feel free to write. You can contact me @

'echo qkdbcsfsb@qpcpy.dpn | tr b-z a-z'

PS - The tentative name of the conference is "Destiny : Linux" but I personally do not think Linux is the answer to everything, and other free operating systems should be discussed and presented as well. And the emphasis on Linux takes emphasis away from the more important point of software freedom. (Heh, maybe I shouldn't worry about getting Stallman to attend, as I see I have his gig covered. ROFL.)
Comment for jdybnis

Could it be the glibc maintainer went on a month's vacation to the Amazon, where he could get away from pestering patch authors? :-) Seriously though, it is likely your patch was lost in the maintainer's daily email shuffle. I am willing to bet the glibc mailing list is not a quiet one. And you're not the only one sending in patches or requests directly to the maintainer. And then there's spam ...

But sometimes patches get ignored, be it because the maintainer was too busy to look at it (see above), or it just didn't excite them enough to be worth the effort of merging it in.

So your patch didn't get accepted on your first try. Putting it up on some random webpage and then forgetting about it just shows the maintainer was right to ignore your patch! Your patch must have been just a quick hack, probably not a lot of effort, for you to just throw it away like that. Who wants that kind of coder as a contributor? I certainly do not!

So if at first you don't succeed ... keep trying. If you care about your patch, and if you believe it is important, keep up with the glibc codebase, and keep resending your patch. Sometimes patches get accepted because the author shows good skills and persistance. Most of the time they are accepted because the author shows good skills, persistance, and good grace.

If you believe you did good work, and that your patch is useful, here is what I recommend you do:

Maintain your patch in sync with the codebase, and kindly (and without a lot of fuss) resend your synced patch to the maintainer every week for four weeks. If after four tries you still do not get a response, post your patch on the development mailing list and ask (kindly) why your patch is being ignored.

Either you will be told flat out what is wrong with it, or it will get accepted. You will learn something either way...

Hacking Gates, Da Man

I suddenly decided some weekends ago that I needed a new file server, immediately. I didn't want to wait for another pay period to buy it, so it had to be really cheap. I would be the only one using it, so it didn't have to be blindingly fast or powerful. But it had to be adequate for NFS and SAMBA file sharing, and for learning Apache, MySQL, and PHP.

So I bought a used XBox, put in a solderless modchip, bought an 80 GB IDE hard disk, and installed Debian GNU/Linux 3.0 r0 with a modified 2.4.20 kernel on "Da Beast." I did not install X or sound drivers on it, and I've configured it to run headless; if / when I need to, I log in through SSH.

Cost of 733 MHz Pentium III, 64 MB RAM, 80 GB HD Linux file server? $308 US

Rise, Phoenix, from your ashes ...

Now that I have plenty of extra storage space on NFS, I can hack again on my old and long-neglected Sega Dreamcast and Broadband Adapter. :-)

I have a Sega Dreamcast with Broadband Adapter, which I bought in February 2001, to hack around with game programming on KOS. It was not getting a lot of play time lately, so I looked into other uses for it. Wonderfully, there are Dreamcast ports of eCos and RedBoot, Linux (here and here), and NetBSD. There are also two homebrew operating systems for Dreamcast: NewOS and KOS, which I already mentioned. NetBSD/Dreamcast and KOS are the most active at the moment.

I burned a CD-R with RedBoot, with which I interact through the Dreamcast's serial port. From RedBoot, I can load eCos, Linux, or NetBSD from NFS. I configured these to run without X and sound drivers. Once loaded, I interact with the OS through the serial port or through SSH.

Note: RedBoot, eCos, Linux, and NetBSD on Dreamcast can be configured to use the TV or VGA adapter output and a Dreamcast keyboard for the console, but I am not interested in that.

My Dreamcast is to be used in learning operating systems programming. Before, when I only had one computer available for hacking, I took things easy because I didn't want to cripple my one and only system. After all, I have other things I want to hack on besides operating systems.

With the Dreamcast, I can experiment and break things all over the place and not worry one bit. I am keeping two copies of each operating system's root filesystem on NFS (that's why the XBox NFS server has an 80 GB hard disk, although it is still overkill). If I break something, I can just reboot from the stable copy, figure out what I did wrong and fix it.

Mono Documentation

In order to continue to contribute to the Mono documentation effort, I am writing a few Gtk# mini-applications, and I am documenting the process through which I am constructing them. This documentation will appear as case studies in the GNOME.NET section of the Mono manual.

One of the mini-applications is, surprise, surprise, a Gtk# rewrite of GNOME Calculator. :-) The version of the code which will be used for Mono's documentation will not use GNOME Calculator's fixed precision math engine. But I will eventually break out GNOME Calculator's math engine into a shared library, and interface with it from the Gtk# GUI with P/Invoke.

Learning Apache, MySQL, and PHP

I bought two O'Reilly PHP books, Programming PHP and Web Database Applications with PHP and MySQL. I devoured these really fast and started to modify the code samples, scripting little test pages to my heart's content. PHP has been the easiest and most fun web server scripting language I've ever worked with. I can definitely see why there are so many PHP-based sites out there.

The purpose of this detour into PHP is two-fold: my free software projects are going to need a homepage of their own if they're going to contribute to free software world domination. I am learning PHP so I can maintain these homepages. The other purpose in learning PHP is that I am helping out a few free software organizations with the upkeep of their web sites, and some of them use PHP extensively.

8 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!