Saturday, July 24. 2004

Well, even though I really rather sit this one out with my baby, I am planning to leave tomarrow to Portland for the 2004 O'Reilly Open Source Convention. Besides being a member of the planning committee for the 4th annual PHP conference at OSCON, I will also be giving two talks. The first is a beginners talk on using forms and PHP, while the second will be an introduction session on writing your own PHP extensions using the C programming language.

If you'll be in Portland next week, I'll see you then!!

Saturday, July 24. 2004

Well, even though I really rather sit this one out with my baby, I am planning to leave tomarrow to Portland for the 2004 O'Reilly Open Source Convention. Besides being a member of the planning committee for the 4th annual PHP conference at OSCON, I will also be giving two talks. The first is a beginners talk on using forms and PHP, while the second will be an introduction session on writing your own PHP extensions using the C programming language.

If you'll be in Portland next week, I'll see you then!!

Wednesday, July 21. 2004

On July 19th, 2004 at 12:17pm I became a incredibly proud father of a 19 inch, 6 lb 9.3 oz baby girl named Diana Katheryn Coggeshall! She's by far the cutest thing on the face of the planet, and both the baby and Hollie are doing wonderfuly. Thank you everyone for your e-mails, comments, text messages, phone calls, flowers, and just general congratulations and support! Both Hollie and I really appreciate it :)

For now this is going to be the only photo I will put online (although I already have tons). Since I'm currently on a dial-up connection I can't afford the bandwidth for more. Not to worry, she'll have plenty of entries in the photo gallery if not her own diana.coggeshall.org section of my web site :)

Update: For those who are interested you can find a constantly updating gallery of pictures of my beautiful girl in the Photo Gallery

Wednesday, July 21. 2004

On July 19th, 2004 at 12:17pm I became a incredibly proud father of a 19 inch, 6 lb 9.3 oz baby girl named Diana Katheryn Coggeshall! She's by far the cutest thing on the face of the planet, and both the baby and Hollie are doing wonderfuly. Thank you everyone for your e-mails, comments, text messages, phone calls, flowers, and just general congratulations and support! Both Hollie and I really appreciate it :)

For now this is going to be the only photo I will put online (although I already have tons). Since I'm currently on a dial-up connection I can't afford the bandwidth for more. Not to worry, she'll have plenty of entries in the photo gallery if not her own diana.coggeshall.org section of my web site :)

Update: For those who are interested you can find a constantly updating gallery of pictures of my beautiful girl in the Photo Gallery

Monday, July 19. 2004

Hollie is in Labor!!!!

I'm on the next flight from New York City to Detroit to be with her!

Wish us luck!

Monday, July 19. 2004

Hollie is in Labor!!!!

I'm on the next flight from New York City to Detroit to be with her!

Wish us luck!

Sunday, July 18. 2004

For those of you who haven't heard yet, Zend is now offering a Certification Test as a PHP Engineer. This 70 or so question standardized test was architected by myself and other members of the PHP Educational Advisory Board -- a group of leaders in the PHP Community charged with defining what exact it means to be a qualified PHP developer.

Besides the test itself there is also an official study guide and php|architect has also just opened a complete certification center on their web site. The certification center will be a great resource for those who are interested in taking the test and will have many resources including inexpensive practice exams. I am currently in the process of writing these practice exams and will be an excellent way to prepare for the real thing.

As for the certifications themselves, you'll have to go to one of the official VUE testing centers. For those of you who will be attending OSCON this year Zend will also have their mobile testing center at the conference for people to get certified. You can Sign up to get tested today and even receive a $50 dollar discount (use the promotional code "OSCON2004") off of the retail price of $200 dollars.

So what are you waiting for? Go get certified! For those organizations which are interested in getting their development staff certified I can arrange to have group discounts so Contact me if that'll be useful to you.

Sunday, July 18. 2004

For those of you who haven't heard yet, Zend is now offering a Certification Test as a PHP Engineer. This 70 or so question standardized test was architected by myself and other members of the PHP Educational Advisory Board -- a group of leaders in the PHP Community charged with defining what exact it means to be a qualified PHP developer.

Besides the test itself there is also an official study guide and php|architect has also just opened a complete certification center on their web site. The certification center will be a great resource for those who are interested in taking the test and will have many resources including inexpensive practice exams. I am currently in the process of writing these practice exams and will be an excellent way to prepare for the real thing.

As for the certifications themselves, you'll have to go to one of the official VUE testing centers. For those of you who will be attending OSCON this year Zend will also have their mobile testing center at the conference for people to get certified. You can Sign up to get tested today and even receive a $50 dollar discount (use the promotional code "OSCON2004") off of the retail price of $200 dollars.

So what are you waiting for? Go get certified! For those organizations which are interested in getting their development staff certified I can arrange to have group discounts so Contact me if that'll be useful to you.

Friday, July 16. 2004

So the other day my Projects page was Slashdotted because of my work attempting to compile PHP scripts into intel assembly code. The project is called PASM, and although I managed to be Somewhat successful in accomplishing my goal I never released any source code for it.

Since Tuesday when the project was made known to the world via Slashdot, I've gotten a flood of e-mails from users asking where they can get a copy of the source tree. Since I am much too busy to really do any signficant work on the project anymore, I figured I'll just release what I have and see if anyone out there is brave enough to make it work. So here you go:

The PASM Source Code

The Project works like this: The 'asm' folder created when you extract the tarball is actually a PHP 5 extension which hooks into Zend Engine 2. That means you'll need PHP 5.0+ or higher to make it work (although I can't even promise you my ext will still compile in PHP 5 -- it was written way before Beta 1 was even released). The purpose of this extension is to basically hook into the Zend compiler and capture the callbacks which occur during script execution for each op-code. Those op-codes are then transfered into a giant associative array which is then accessiable within PHP itself. The second half of this project is the actual ASM code generation and it consists of three distinct parts. The first part is the PASM compiler framework, which is a PHP script in the pasm directory called pasm. It's job is to use the extension mentioned previously to call a code generation module that actually generates the assembler code. That module is the x86 module and it is found in x86_module.inc. This is where the ASM code is actually produced. The second part of the code generation is the library of FASM assembler marcos I created found in files such as pasm_var_macros.inc. These assembler routines are designed to provide "typeless" variables and on-the-fly conversion of them between types to mirror the behavior in PHP. The third and final aspect, which perhaps is the most stable of all of the things this project has done, is the PMM (or PASM memory manager). This is a completely self-contained modular memory manager written enitrely in FASM assembler. There are does (MM_README) for it so you can go from there.

To compile anything, you'll need the FASM compiler and GCC. You can find FASM right here. If you don't already have GCC, you might as well not even bother with this one.

So there you go. Download it, be aware it is likely not to work, and considering it has been the first and only attempt of its kind I don't want any hate mail saying what a bad job I did with it. If you somehow find it useful, or even better thing you can fix it -- let me know :)

Friday, July 16. 2004

So the other day my Projects page was Slashdotted because of my work attempting to compile PHP scripts into intel assembly code. The project is called PASM, and although I managed to be Somewhat successful in accomplishing my goal I never released any source code for it.

Since Tuesday when the project was made known to the world via Slashdot, I've gotten a flood of e-mails from users asking where they can get a copy of the source tree. Since I am much too busy to really do any signficant work on the project anymore, I figured I'll just release what I have and see if anyone out there is brave enough to make it work. So here you go:

The PASM Source Code

The Project works like this: The 'asm' folder created when you extract the tarball is actually a PHP 5 extension which hooks into Zend Engine 2. That means you'll need PHP 5.0+ or higher to make it work (although I can't even promise you my ext will still compile in PHP 5 -- it was written way before Beta 1 was even released). The purpose of this extension is to basically hook into the Zend compiler and capture the callbacks which occur during script execution for each op-code. Those op-codes are then transfered into a giant associative array which is then accessiable within PHP itself. The second half of this project is the actual ASM code generation and it consists of three distinct parts. The first part is the PASM compiler framework, which is a PHP script in the pasm directory called pasm. It's job is to use the extension mentioned previously to call a code generation module that actually generates the assembler code. That module is the x86 module and it is found in x86_module.inc. This is where the ASM code is actually produced. The second part of the code generation is the library of FASM assembler marcos I created found in files such as pasm_var_macros.inc. These assembler routines are designed to provide "typeless" variables and on-the-fly conversion of them between types to mirror the behavior in PHP. The third and final aspect, which perhaps is the most stable of all of the things this project has done, is the PMM (or PASM memory manager). This is a completely self-contained modular memory manager written enitrely in FASM assembler. There are does (MM_README) for it so you can go from there.

To compile anything, you'll need the FASM compiler and GCC. You can find FASM right here. If you don't already have GCC, you might as well not even bother with this one.

So there you go. Download it, be aware it is likely not to work, and considering it has been the first and only attempt of its kind I don't want any hate mail saying what a bad job I did with it. If you somehow find it useful, or even better thing you can fix it -- let me know :)

Saturday, July 3. 2004

While I was at LinuxTag, I was a participant in a protest against software patents (you can see some pictures of our march, etc. in my LinuxTag 2004 Photo Album). Today I got an e-mail from its organizer informing me of a second protest in Germany and gave me a link to some photos taken during the last protest. You can find those photos here.

Besides being very wonderful photos, looking through them got me thinking about the topic of software patents in a little more detail. Rather than pondering them myself I thought I'd share them with my faithful blog readers -- please feel free to comment and share your thoughts as well.

To be honest, I'm not against software patents -- at least not accross the board. I'm a business man working in an open-source industry, I have a family on the way and I can't put food on the table giving everything I do away. I do agree that software patents can be a very negative thing, however the reasons why they are a bad thing aren't necessarily the same reasons as many people who protested with me believed. Many who are completely against software patents claim that they stunt growth in the industry, allow the formation of monopolies, etc. I say that software patents allow me to, if I choose, actually create a piece software or a better way to do things which I can claim as mine. I don't need to be employed with a major company to patent something, and with software I don't even really need a lot of resources to write "patentable" software. So with software being so easy to reverse engineer, what's so wrong with the idea of there being some sort of assignment of credit for the original idea?

To me, the problem doesn't seem to be that patents on software themselves are bad -- rather that people are trying to patent things which shouldn't be patentable. For instance, consider the progress bar. In Europe the progress bar is patented by someone (although I don't know who). I'm against that, not because its a software patent, but because at least in America you aren't allowed to patent "common knowledge" things. For instance, I can't run off and patent the wheel -- and in today's day and age the progress bar is just as common. On the other hand if I (or anyone, for that matter) develops a new compression algorithm which can compress any data 50% better than any other algorithm, why can't they patent that? It's something they completely thought up on their own, and I think if you develop something unique enough you should have a right to be the first (and only for a time) to gain the benfits of it.

Patents aren't the problem, its the stupid laws that govern the way patents work. You shouldn't be able to patent a piece of software that is considered "common knowledge" -- but when someone comes up with something new, I think they deserve the first bite at the apple for it. If that company is Microsoft, so be it -- you can bet they spent the money to develop it. But the great thing about our industry is that although Microsoft can throw millions to solve a problem, a clever guy can sit in front of his laptop at a bar and come up with the same thing. To a large degree, that's the power of open source -- and when you look at it that way the idea of patenting something actually seems to favor the individual shoebox developer anyway. Unfortunately, the thing about open source is that there is a real danger from some who seem to be content int giving ourselves away into non-existance. Sooner or later everyone needs to make money off their talents, and while open source is an excellent way to prove yourself while you improve the entire industry, writing open source code alone isn't going to put food on anyone's table. Patents are okay, patents which are completely based on the work of others with a slight modification, or patents which are based on technologies and ideas which are common knowledge, are not.

At least that's how I look at it.. let's hear what you think.

Saturday, July 3. 2004

While I was at LinuxTag, I was a participant in a protest against software patents (you can see some pictures of our march, etc. in my LinuxTag 2004 Photo Album). Today I got an e-mail from its organizer informing me of a second protest in Germany and gave me a link to some photos taken during the last protest. You can find those photos here.

Besides being very wonderful photos, looking through them got me thinking about the topic of software patents in a little more detail. Rather than pondering them myself I thought I'd share them with my faithful blog readers -- please feel free to comment and share your thoughts as well.

To be honest, I'm not against software patents -- at least not accross the board. I'm a business man working in an open-source industry, I have a family on the way and I can't put food on the table giving everything I do away. I do agree that software patents can be a very negative thing, however the reasons why they are a bad thing aren't necessarily the same reasons as many people who protested with me believed. Many who are completely against software patents claim that they stunt growth in the industry, allow the formation of monopolies, etc. I say that software patents allow me to, if I choose, actually create a piece software or a better way to do things which I can claim as mine. I don't need to be employed with a major company to patent something, and with software I don't even really need a lot of resources to write "patentable" software. So with software being so easy to reverse engineer, what's so wrong with the idea of there being some sort of assignment of credit for the original idea?

To me, the problem doesn't seem to be that patents on software themselves are bad -- rather that people are trying to patent things which shouldn't be patentable. For instance, consider the progress bar. In Europe the progress bar is patented by someone (although I don't know who). I'm against that, not because its a software patent, but because at least in America you aren't allowed to patent "common knowledge" things. For instance, I can't run off and patent the wheel -- and in today's day and age the progress bar is just as common. On the other hand if I (or anyone, for that matter) develops a new compression algorithm which can compress any data 50% better than any other algorithm, why can't they patent that? It's something they completely thought up on their own, and I think if you develop something unique enough you should have a right to be the first (and only for a time) to gain the benfits of it.

Patents aren't the problem, its the stupid laws that govern the way patents work. You shouldn't be able to patent a piece of software that is considered "common knowledge" -- but when someone comes up with something new, I think they deserve the first bite at the apple for it. If that company is Microsoft, so be it -- you can bet they spent the money to develop it. But the great thing about our industry is that although Microsoft can throw millions to solve a problem, a clever guy can sit in front of his laptop at a bar and come up with the same thing. To a large degree, that's the power of open source -- and when you look at it that way the idea of patenting something actually seems to favor the individual shoebox developer anyway. Unfortunately, the thing about open source is that there is a real danger from some who seem to be content int giving ourselves away into non-existance. Sooner or later everyone needs to make money off their talents, and while open source is an excellent way to prove yourself while you improve the entire industry, writing open source code alone isn't going to put food on anyone's table. Patents are okay, patents which are completely based on the work of others with a slight modification, or patents which are based on technologies and ideas which are common knowledge, are not.

At least that's how I look at it.. let's hear what you think.

Friday, July 2. 2004

Well, I've moved my life to New York City and am in the process of finding a place to call home for the next few years. However, I did want to take a moment and wish all of those Americans who read my blog a happy 4th of July weekend! I'd wait until the actual day, but I'm not sure I'll be online then.

Be safe, enjoy the fireworks!