Registration is open for ApacheCon Europe. We have a good schedule of talks, and some interesting speakers. Should be a good time.
Tag Archives: apache
Waiting for Godot
I renamed the vhost on my laptop …
Well, it made me giggle, anyways.
Axis2 Feathercast
There’s a new feathercast, and I’m very pleased with it, because this is the first of the feathercasts that’s really the type of thing I wanted to see FeatherCast.org doing in the first place. (With the possible exception of the Harmony episode, I suppose.)
Much of the motivation for doing FeatherCast in the first place was to answer my own questions about all of those Apache projects that I hear about, which I have no idea what they do.
Clearly, the Axis2 team has a lot of energy and a lot of brilliant people. But until ApacheCon Asia, I had not the vaguest idea of what the software actually did. Now I have a general idea of what it does, and how the other projects like Sandesha fit into it, and how they got their strange names. This, in turn, makes me understand how Axis2 fits into the larger ASF.
Eventually, one hopes that we can understand what all the various Apache projects do, and we can be more than a collection of disconnected factions, but can actually be a federation of related nations.
On a related note, if you’re going to be at ApacheCon, and you work on one of those projects that always generates the comment “That’s an Apache project??!? What does it do??”, please please please find David Reid or myself and record a feathercast about your project. It benefits the entire ASF, but it most benefits your project when the rest of us know what niche you fill in the greater ecology.
20 Things You Didn’t Know Apache Could Do
In just a couple weeks I’ll be giving a talk at Ohio LinuxFest titled “20 Things You Didn’t Know Your Apache Web Server Could Do.”
I picked the number 20 mostly at random. I figured that was a good number for a 45 minute presentation. It turns out it’s hard to come up with 20 things that I think most of the audience at OLF won’t know. They tend to be a pretty knowledgeable audience.
Tonight I came up with the last few things, and I think I’m mostly ready to give the talk, although some of the items require a little more detail. Not too much, because 2 minutes per item is pretty tight.
I figure that most of the audience will find 2-5 of the things to be something that they already knew, and that a few very experienced folks will perhaps know as many as 10 of the things, but I’m betting that nobody in the room will know more than 12. We’ll see. It will be a fun talk, whatever happens.
Oh, yeah, I’ll be giving the same talk at ApacheCon in Austin. Rather different audience. I imagine there will be a person or two in the audience there who will know all 20 things. But perhaps I’ll even teach them a few things.
Perl stuff in svn
For many years, I worked on Perl date/time modules. Calendars – non-Gregorian ones, that is – held a great deal of fascination for me, and I wrote Perl modules for many of them. They’re all on CPAN, along with assorted other stuff.
Then, due to a variety of factors, mostly involving Apache, I gradually lost interest in writing Perl calendar modules, and those modules languished. At the same time, the great Perl DateTime module project started (at least in part because of my work, but mostly due to Dave Rolsky) and most of my modules became obsolete, and replaced with the better implementations that came out of that project.
I just rediscovered all my Perl date/time modules in cvs, as I was moving all CVS stuff over to svn. here they are, if anyone cares.
There’s some other stuff in there too, if anyone wants to poke around. If you want to submit patches, please feel free to do so. If you want to fork any of these projects and do something else with them, please feel free to do that, too, although I’d kinda like to know about it if they amount to anything useful.
Everything that you can get to in the public/ directory should be considered to be under ASL2 (Apache Software License version 2.0) unless explicitly stated otherwise.
AuthUserFile /dev/null
By the way, I mentioned sarcastically a few days ago that there are 48,000 howtos with incorrect advice about HTTP authentication.
Turns out I was wrong. There’s roughly 75,000. And about 60,000 that tell you to use AuthGroupFile /dev/null.
Now, I miss 1995 as much as the next guy, but, alas, it’s time to move on.
FeatherCast Swag
You should go buy some FeatherCast Swag.
10G of audio
By the end of last week, I had more than 10G of audio on my desktop, from various interviews that I had done and not yet edited. This weekend, I got rid of about 3G of that, but there’s still a lot to go. So hopefully there will be new FeatherCast episodes in the next few days. Sounds like David has been recording some stuff too. Not sure which one(s) will get released first, but with ApacheCon US coming up real soon, there’s less need to ration them.
The future of (Open Source) software
It’s easy for us Open Source “experts” to go to various places around the world and impart our wisdom on the people there. But what I realized in the past week is that it’s really only a matter of time before they are telling us the best way to do Open Source. Not to mention that it’s just the wrong attitude to take, since it’s only when everyone are equal participants that the Open Source model really works.
Anyways, I need to stew on this a little longer, because there’s a lot to think about. Mostly, though, I tend to have rather negative reactions to any model that has “us” from the west teaching “them” from the developing world what the right way to do things is.
I finally figured out what WS02 is all about, and it seems that they grasp what Open Source is meant to be as a business model. They are a software company, and their product is Open Source. So they benefit from the contributions of folks around the world, and those folks benefit in return, but they are still able to run a profitable software company based on those efforts.
It occurred to me to make the (very unfair, but stick with me) comparison to ISPs in the USA who have based a business around the Apache web server. There are dozens of them. They sell services on top of an Open Source product, but in no way contribute back to the community that makes their business possible. Quite to the contrary, they do things that subtly work against it. My biggest pet peeve in this area is ISPs who have “documentation” on their website about how to run your hosted website. Granted, *some* of it is worthwhile, and some of it even links back to the relevant parts of the real documentation. But for the most part, ISP Apache documentation is full of inaccuracies, stuff that stopped being true in 1997, confusing phrasing, and misconceptions about best practice.
If we could get all of these documentation writers to participate in some small way in the actual Apache web server documentation project, contributing howtos and documentation patches, as well as correcting their own misconceptions, imagine the wealth of howtos and documentation we would have now.
But, instead, we have 48,000 howtos that tell you that authentication requires .htaccess files and “Limit GET POST” blocks. The sheer enormity of wasted effort staggers the mind.
Hmm. I seem to have gotten slightly off track. I was talking about the future of software.
Anyways, think about this. India is making a name for itself as a place to outsource development and support. So companies in the USA send their work to India, but the vision and plan is still coming out of the USA. Sri Lanka, on the other hand, is making a name for itself in Open Source software. Sri Lanka has a higher per-capita rate of Apache developers than any other nation on earth. And, Sri Lankans are driving the vision and plan for the entire Web Services suite of projects, in addition to participating in other projects.
In an industry where geography is uninportant, we’re going to see more and more powerhouse software companies coming out of developing nations where they grok Open Source as a business model in ways that we in the west don’t yet. I’m obviously not the first person to make this observation, but having see it first hand, it seems much more real to me. Sri Lanka and Brazil are the places to watch.
Perl lightning talks
The Perl lightning talks have been a staple of the Perl Conference, and, later, the O’Reilly Open Source Convention, for as long as I’ve been attending it. And YAPC, too, although I haven’t been to YAPC in many moons. They were somewhat different this year. Not hugely, but subtly.
MJD observed that, when he started them, the purpose was to get folks who would otherwise not give talks to give brief presentations in a low-stress environment. 5 minutes is enough to get a taste of public presenting, but not long enough to get too terribly intimidated.
Over the years, it became a bit of a sideshow, with elaborate presentations, complete with slides and sound effects. And of course MJD as the MC, complete with funny hat and gong. And so folks who wanted to give a five minute “here’s what I’m working on” or “here’s my cool idea” talk were overshadowed by the brilliant presentations by Audrey Tang and Andy Lester and the like. Also, talks lean towards the comic routine rather than the technical talk. Indeed, technical lightning talks tend to get heckled on IRC, and yawned through, waiting for the *real* lightning talks.
We started doing Lightning Talks at ApacheCon a few years back, and they have become part of our conference culture. We, too, tend to favor the entertainment talks rather than the technical talks. That’s fine, in that it draws a crowd, and folks hear interesting ideas, and it’s a great community event. But we need to remember what the initial purpose of lightning talks really is – to give folks a shot at the mic, if only for a few minutes.
Of course, I don’t run the lightning talks, so feel free to ignore me. 🙂