mcpantz.org

I just requested a CPAN account last night.  Hopefully I’ll have the gumption to throw up all those that I feel are missing that I’ve been missing that I would be very happy to add.  one of them is glob-match support in perl that’s file-system agnostic.

Of course, My first module’s going to be a result of some Scav Hunt tomfoolery, which I’ll be sure to post about later, once it’s up.  Let’s just say that Jewish numerology will get a lot easier now.

What did i do this weekend?  I helped build a loft, that’s what.  And it was awesome.

Next up on the agenda: I feel sometimes that I’m the leader of the software interests at PS:1 somewhat, so I’m going to make an executive decision on this.  I have many, many different projects going on aside from hacking, so I can’t show up at the space every day like some of the more dedicated members have been doing.  Instead, I’m just going to be regular about this.

I’m going to try and make it to the space Wednesday night and Sunday afternoon.  Anyone interested in programming languages, esoteric languages, functional programming, and software engineering should show up and talk/yell at me.  If enough people show up, we’ll call it a meetup group.  If they don’t, I’ll read & study by myself, and get anti-social and even more stabbity.  Your call.

How do you like having an extremely extensible, mature, and stable text editor. I sure do. One thing that always bugged me when writing code is going through and finding all sorts of annoying trailing whitespace. For most major modes, tabs are converted to spaces for consistency, but trailing whitespaces are neglected. I was considering writing an elisp function to go ahead and take care this until I found This little gem. I suggest that anyone that has to deal with syntax where trailing whitespace is ignored to have a crack at this.

For that matter, the rest of the interface enhancements on the main page look like they’d be excellent additions to one’s ~/elisp/ folder even today; as for the most part, the code here dates back to the early 90’s. Emacs isn’t exactly a popular mail client anymore…

Google Labs just released Lively, a supposed contender to Second Life. I played the latter a decent amount to see what it has become, which is mostly a desperate cry for more and more attention. People will wander about looking for someone to talk to, and run into a variety of weirdos. I predict a combination of late 90’s AOL chatrooms, IRC gaming, or any other effort at suspended reality that just feels empty and unfulfilling.

The general cycle will probably engender the following:

• early technology adopters and cool-hunters enter the scene, encountering a vast wasteland; declare it simultaneously the next best thing and a complete waste of time
• tweens, internet-based subcultures, and perverts invade the start inhabiting the wasteland. The first page is already filled with sex rooms.
• the technology languishes on in limbo with Google occasionally trying to inject life into the project, only to be met with general disinterest and all-around frustration.

I should probably note that I haven’t actually tried Lively yet. There are a number of things that Lively does get right, and should probably be adapted by any future virtual worlds:

• Separate rooms. Mind you, they could have called it something else that doesn’t make it feel like IRC, but it’s the right idea. Keeping every area of a persistent world totally active in a distributed environment sucks resources that may *never* be used.it makes much more sense to just enforce a boundary & throw up a skybox on users and make them talk to each other.
• Browser integration. No one wants to boot up a full-screen app to just see what rooms are popular and maybe if your friends are online.
• An offline editor. Rumors persist that Sketchup 7 may very well allow people to design objects in a dedicated atmosphere instead of trying to do so in-game, allowing people to focus instead of deal with a bunch of furries running by in morally repulsive costumes.

However, there are some other areas where it falls flat on its face:

• Lack of user content from the get-go. Google’s Don’t Announce Anything policy does give them the freedom to drop something When It’s Ready, and the lack of user content is just going to result in a bunchor restless users who are tired of seeing the same fucking samuri squirrel run around. On another level, it does give them the ability to establish a look and feel to the program, but users may find that more stifling in the end.
• Platform lock-in. I know, I know, most of the civilized world uses windows, but most of the world who is going to write reviews of the software does not. Especially design professionals; the ones who will actually be the cogs in the virtual market they are so desperately trying to create.

The real problem is that no one’s found a good reason to use virtual world technology in a non-gaming context. Second Life general offensiveness in getting *anything* done in a reasonable time in-world has turned off a number of potential users, including myself.  However, there is a lot of potential markets that a decent implementation could access.  It’s that latter part that really means something.

In the meantime however, I’m just going to curse until someone creates a virtual world that doesn’t completely mock the cyberpunk ideal.

I have been fighting with Samba recently, to get it configured correctly. I only wish this wasn’t such a black art, but it’s what I do. I had thought backups were working, but something far more sinister was happening, and I didn’t know what to do about it. A coworker of mine recently suggested that we hang out in the #amanda room on irc.freenode.net, so I hopped over to #samba. one half hour later; I learned a lot more than I thought I could in a chat room.

So here’s a hint for all you young programmers out there: open source is here to help, and a lot of us are bored and hanging out on IRC, because we are cool check it out! In fact, if anyone would like to talk about BackupPC, join me on irc.freenode.net in #backuppc.

If I haven’t already given you an ecstatic phone call or instant message, then I should tell you that I’ve just accepted an offer from Zmanda as a design engineer. I feel really lucky with this; as I’m going to be working in my most familiar language (Perl) to make some fairly robust code user-friendly. I’ll be heading up a new project to investigate adding a new open-source project to our lineup; so I’ll be in a lot of murky waters, but it’s going to be a lot of fun to work with, and give I’ll have the opportunity to collaborate while being fairly autonomous.

I’m sure I’ll be blogging more about the crazy things I’m going to end up doing in Perl as a result. Object-oriented Perl is simple yet quite bizarre, so it’ll be an interesting experience to delve further into its depths. I also have a couple fun ideas for some perl modules, so keep your eyes open.

A gem of a post came across my alma mater’s computer science mailing list: Prime detection using regular expressions. It’s quite a clever hack; what it actually defines is all numbers that are not prime, but the negation of that creates the primes.

It’s that pesky /\1+/ that makes this whole expression work, and it’s what prevents this regex from being a regular expression. Regexes have long since stopped being regular expressions, but it’s a programming language unto itself.

A while back, I did LaTeX parsing in Perl, and used a one-level parser (custom for LaTeX) and a incredible amount of regexes in order so slice and dice this massive corpus of data. Unfortunately, I didn’t know about Parse::RecDescent then, or else I would have put myself out of a job very quickly. But I wonder, can parentheses (a recursive language) be validated as a regular expression language yet? Better yet, is the language itself Turing complete? Perl regexes, at least, have a lot of very extended features, and this will require some unusual digging.