I hated to do it, but lets be realistic. Its already September the 28th, and so far this month I’ve logged about 10 total hours on the development of DiyShare. There is probably more like 80 more hours required, if I had to estimate it, so I am pushing the launch date back to Halloween. Basically I’m giving myself the month of October to complete the development and testing. There is already something like 50 or so hours already invested in the development, so a good chunk of work has already been done, and while I am discouraged with having to do this, I am not overly so, because of the work that’s already been put into the site.
September has just been an insanely busy month for us. Lots of things going on that has required our time and that means that Diysearch and its associated projects takes a back seat. There are still a number of features I need to add for our Editor that I haven’t gotten to yet. Juggling the priorities is difficult because, while our editor needs the tools to effectively do her job, they have to be weighed with getting DiyShare finished. Often, tools for Diysearch have taken a priority in that Diysearch is a production system, its live and is actively being used, so that means a new system (DiyShare) gets pushed back.
It’ll get there. A little less time with the Warcraft and in 5 weeks it should be ready.
I suppose it’s my own fault. Diysearch used to be in the top 3-5 results in google for the keyword: “DIY.” But now? Top of the second page, number 11. Grrr. I’m frustrated. I don’t even know why, this stupid project was never a popularity contest for me, but with the traffic to the site not increasing and in fact, continuing to slide, I’m just a bit stumped as to what is going on. Actually, I have a gut feeling I know what is going on.
Here’s my take. Diysearch has, in the eyes of many folks, especially new users, become irrelevant. I mean, why bother using a site like this when you’ve got Google. So, what I think I need to do is, make the case of why you should be using Diysearch and not Google. There are solid reasons, mostly having to do with control. With Diysearch control is put into the user’s hand. Links, relevance, ratings, all of that stuff is controlled by users. With Google you wait, hoping your site gets indexed and then hope you’ve done your homework right so that you don’t get burried into oblivion and hope that someone *else* thinks you’ve done your homework right so as to boost your page rank. No, here things are done differently. Here, you determin where you get listed, how you get listed and relevance is determined not by someone else, but by you. We can do that, because we are very particular who gets in in the first place… which while is counter to the “user in control” concept, actually better facilitates this through a selective process that ensures crap doesn’t end up polluting everything else.
Yeah, I do not have expectations of being serious competition to Google (duh), no in fact my expectations are rather pedestrian, I just want to be useful. I’ve invested nearly 20 years of my life in this DIY thing, I just want to make a difference.
I am reticent about filing this under “site news” because in reality it kinda isn’t. But in some weird tangential way it is. This is also me just trying to keep myself motivated and not let this thing get stale (as I often have a habit of doing), but the sad truth is, I don’t have a lot to report.
Yes, I’ve been doing some work here and there on the various new category functions (like “watch” feature, which is going to be really cool when I figure out how to make it so that any “object” in the site could be “watchable,” because the last thing I want to do is paint myself into a corner of hardcoding functionality like that) and the next big thing will be implementing a content management system for the non-database driven pages.
Our editor is always looking for opportunities to better present content within the site, and currently “static” pages are encapsulated as templates with embedded XHTML (for presentation). Well, this makes it difficult to edit. I decided on OpenCMS as the content management system. OpenCMS is a very mature, enterprise-class content management system more like an Interwoven TeamSite than say something like a Droopal or Joomla (which I didn’t like because if you choose one of those products, you are forced into doing things their way. I don’t like that). With OpenCMS I can define content types and elements using standard XML schema, and then transform XML-based content using JSP (Java Server Pages) into the presentation templates currently used within the site. So publishing is quite simple. There are a few odds and ends I need to finish up (like a batch job to push the managed content into the UAT and production sites).
What else? Oh yeah, development on DiyShare is grindingly slow, which brings me to the title of this entry. Instead of spending all of my waking spare time writing code, I’ve been playing the Warcraft. Actually I’ve been playing this game on and off for about 9 months, and have been sinking enormous amounts of time in leveling my main (a night elf druid, who is currently level 45, nearly 46). So, if you play and happen to play on Gilneas and happen across XBlinder, that’s me. Come tell me to stop with the Warcraft and get back to working on the site!
Well, the server move is all done, and everything is up and ready. Aside from the fact that I completely slept in and our editor let me know by asking “uh, do you know the site’s down?” Something had happened in one of the more obscure apache settings (that I wasn’t even using at the time) that wasn’t letting the HTTP server from coming up completely, so it was unable to server requests. That’s why if you were trying to look at the site earlier this morning you couldn’t get the darn page to load. Sorry about that… my fault.
Anyway, the move is complete, we are on the new “phat pipe” and hopefully you should notice an improvement in performance (not that performance was bad, because it wasn’t) but improvements are always good.
Next time something like this happens, I’ll set an alarm and get my lazy bum out of bed to make sure everything is working right, immediately, now 4 hours after the fact.
