While we're not exactly Uncle Sam, SIX APART NEEDS YOU to come train at Camp Trott and join us as we fight for the right to publish blogs and rearrange widgets. And by that, I mean attend our next Movable Type Open Source hackathon and build the kind of cool stuff that made you fall in love with blogging and coding in the first place.
However, first there are some minor tactical details we need to settle on. We are thinking of some time in early April, but before setting the
exact date we'd like to know whether a workday or weekend day works
better for those of you in the community.
If the hackathon is held during a workday, it would probably begin in late morning, and continue on into the night, so you could show up after you finish your day job, if you choose. A weekend hackathon might take place all day Saturday or Sunday; or even, if there is enough interest, on both days. No matter when it is held, though, rest assured, you will be adequately supplied with everything you need to fight, and fight well (ie, free sodas, coffee, tea, salty snacks, wired and un-wired internet,) Freedom will also be supplied.
So please comment and let us know your thoughts on when you'd like to join the cause.
Are you proposing a hackathon or a sprint? A code sprint would bring together MTOS developers to commit patches to 4.2 to achieve specific feature goals (performance in this case). Of course it would be nice to have commit access first. A hackathon might instead focus on the extensibility of MTOS, encouraging new plugins, extending the rich text engine, or creating new themes.
Scope please! I prefer a MTOS 4.2 sprint focused on performance. Perhaps we could put a little more open source mentality in MTOS by that target date.
Sounds like fun, but unfortunately y’all are on the opposite side of the country ;)
If I might make a suggestion, pass this around to more web designers than developers. What Movable Type really needs now is to get is a boat load of themes. WordPress has what, 2,000 or more of them now for free? Time to show them that they’re not the only game in town!
They have some interesting performance gripes. I don’t think the complaint about the performance issues related to the plugin are valid, but it would definitely be a good idea to make the focus of 4.2’s performance improvements be on rebuilding.
I don’t know much about Perl (MT is the extent of my use of Perl), but I have a few suggestions on where performance can be improved:
1) Make a BootStrapLite package, that exposes just the ORM functionality and provide clean processing of POST/GET variables. I’ve had cases where all I wanted to do was take in some request data, wrap them into objects, and save them. Can’t do that in a lightweight way from what I’ve seen.
2) Make the installer able to autodetect modperl and configure the trackback and comment scripts for modperl if possible. This is where WordPress gets the illusion that it’s significantly faster than Movable Type. MT has to run as a CGI app by default, whereas WordPress gets the benefit of mod_php.
3) By default split the rebuild process into two phases. Have the individual archive and main index updated immediately on rebuild, but place the rebuild of the other archives into a delayed publishing queue that gets updated every hour.
4) Get rid of the author archives mapping by default. This is just waste for most bloggers. In fact, I think category templates ought to be removed by default as well. Why not go crazy here, and have the next release of MT, by default, not do categories on a new install? Tagging is superior to categories.
@MikeT - #4 Author archives are currently disabled in all new installs starting with MT4.2. Removing Categories sure would simplify and speed a default install even more. =)
I agree with you about themes, how do you envision this working?
Here’s some options possible options:
A - one template set, many themes (a la classic MT)
B - 1-to-1 template set to theme ratio (a la wordpress)
C - many template sets, many themes for each template set
I like both the one-to-many and one-to-one options. I’ve got a few WordPress themes that I’ve converted here, that are based on the one-to-many approach.
Everyone,
http://mt-hacks.com/cacheblock.html is a great idea. This is one of the areas where WordPress, through plugins, has gained a performance increase in recent years. Movable Type would benefit greatly from having this sort of functionality integrated into its rebuild system.
I’m definitely interested! Would prefer a weekday, weekend day or weekend night in that order. Also, either a hackathon or sprint would be fine with me.
Arvind Satyanarayan on March 14, 2008, 2:22 p.m. Reply
I vote weekend because I’m more likely to make it then :)
Niall Kennedy on March 14, 2008, 3:10 p.m. Reply
Are you proposing a hackathon or a sprint? A code sprint would bring together MTOS developers to commit patches to 4.2 to achieve specific feature goals (performance in this case). Of course it would be nice to have commit access first. A hackathon might instead focus on the extensibility of MTOS, encouraging new plugins, extending the rich text engine, or creating new themes.
Scope please! I prefer a MTOS 4.2 sprint focused on performance. Perhaps we could put a little more open source mentality in MTOS by that target date.
MikeT on March 14, 2008, 3:47 p.m. Reply
Sounds like fun, but unfortunately y’all are on the opposite side of the country ;)
If I might make a suggestion, pass this around to more web designers than developers. What Movable Type really needs now is to get is a boat load of themes. WordPress has what, 2,000 or more of them now for free? Time to show them that they’re not the only game in town!
MikeT on March 14, 2008, 4:05 p.m. Reply
I have noooo idea :)
Bryan Tighe on March 14, 2008, 5:08 p.m. Reply
A design contest, with prizes!
MikeT on March 14, 2008, 6:03 p.m. Reply
http://birdhouse.org/blog/2008/02/07/notes-on-a-massive-wordpress-migration/
They have some interesting performance gripes. I don’t think the complaint about the performance issues related to the plugin are valid, but it would definitely be a good idea to make the focus of 4.2’s performance improvements be on rebuilding.
I don’t know much about Perl (MT is the extent of my use of Perl), but I have a few suggestions on where performance can be improved:
1) Make a BootStrapLite package, that exposes just the ORM functionality and provide clean processing of POST/GET variables. I’ve had cases where all I wanted to do was take in some request data, wrap them into objects, and save them. Can’t do that in a lightweight way from what I’ve seen.
2) Make the installer able to autodetect modperl and configure the trackback and comment scripts for modperl if possible. This is where WordPress gets the illusion that it’s significantly faster than Movable Type. MT has to run as a CGI app by default, whereas WordPress gets the benefit of mod_php.
3) By default split the rebuild process into two phases. Have the individual archive and main index updated immediately on rebuild, but place the rebuild of the other archives into a delayed publishing queue that gets updated every hour.
4) Get rid of the author archives mapping by default. This is just waste for most bloggers. In fact, I think category templates ought to be removed by default as well. Why not go crazy here, and have the next release of MT, by default, not do categories on a new install? Tagging is superior to categories.
Beau Smith on March 14, 2008, 7:13 p.m. Reply
@MikeT - #4 Author archives are currently disabled in all new installs starting with MT4.2. Removing Categories sure would simplify and speed a default install even more. =)
I agree with you about themes, how do you envision this working?
Here’s some options possible options:
A - one template set, many themes (a la classic MT) B - 1-to-1 template set to theme ratio (a la wordpress) C - many template sets, many themes for each template set
Fumiaki Yoshimatsu on March 14, 2008, 7:43 p.m. Reply
Either it’s workday or weekend, if it’s held in Matterhorn the conf room, we’ll be joining from Tokyo via video conferencing.
MikeT on March 15, 2008, 7:26 a.m. Reply
Beau,
I like both the one-to-many and one-to-one options. I’ve got a few WordPress themes that I’ve converted here, that are based on the one-to-many approach.
Everyone,
http://mt-hacks.com/cacheblock.html is a great idea. This is one of the areas where WordPress, through plugins, has gained a performance increase in recent years. Movable Type would benefit greatly from having this sort of functionality integrated into its rebuild system.
Jay Allen on March 19, 2008, 6:20 p.m. Reply
I’m definitely interested! Would prefer a weekday, weekend day or weekend night in that order. Also, either a hackathon or sprint would be fine with me.
Arelav on March 22, 2008, 1:10 a.m. Reply
I wish I could make it but I am slammed at work, but hope you get a lot accomplished when you have it.
Ms. Jen on March 25, 2008, 3:26 p.m. Reply
Hola,
Excellent idea. I vote for a weekend day or if a weekday one next to the weekend, so that I can come on up to SF.
I also vote to get more designers involved. Let the ProNet folks know about it.
smiles, jen ;o)
Pillowfight on March 26, 2008, 1:07 p.m. Reply
My vote is weekends!
Antuny on July 2, 2012, 2:56 a.m. Reply
Ooooh, I’m so bummed I missed it!! Is there going to
be something like that anytime soon?
mazapoint on July 29, 2012, 9:20 p.m. Reply
A weekend hackathon strength obtain rest all daylight hours Saturday or Sunday or smooth, if here is enough curiosity, on both existence.
Viktor Carembew on August 6, 2012, 6:00 a.m. Reply
I am in without doubth, but are you still doing such things these days?
pro solution on September 5, 2012, 3:18 a.m. Reply
There was something similar to this for Wordpress that I attended. It was great to meet up with people and share ideas.
bahria town on October 13, 2012, 11:50 a.m. Reply
hard work you have made in writing this post. I am hopeful the same best work from you in the upcoming as well