Half-Elf on Tech

Thoughts From a Professional Lesbian

Tag: cms

  • I Haven’t Got Time For The Pain!

    Carly Simon and you should get the joke here Two months ago (give or take) I mused over photo gallery options for my sites. For Ipstenu, I’m now using WordPress and treating it like a photoblog. For JFO, however, I couldn’t answer it that easily.

    I really do like the Gallery project. I do! I learned a great deal about photography from it, and I’m thankful for it. But. I needed to move on as a user, a developer and a photographer. On that last one, I’m not a profession one, I’m just a goofy girl with a camera who likes to remember where she’s been. As a user, Gallery2 did the job well and without major issues. As a developer, it made me want to cry. Many times. Once I had to log into my friend’s server to fix his install. That just whomps.

    Even the developers admit that Gallery2 suffered from bloat:

    The code base is too complex and over-engineered because it was designed to fix every single thing that was wrong with Gallery 1 (Second System Effect) leaving its scope hazy and broad.

    The whole idea of it was “Your photos, your website.” And personally I love that. I hate having flikr or picasa in charge of MY photos. Let alone FaceBook. I have a blog on my domain for that same reason. But Gallery2 was too much. I never used half of it and it was 16+megs at its slimmest install. That the developers agreed with my feelings delighted me. And the Feature List was also exciting. As soon as G3 popped out, I grabbed a copy and started playing.

    With each version of Gallery3’s beta releases, I would get excited and then disappointed. Excited for the new toys and disappointed for how the overall effect felt. It just felt wrong for me. It wasn’t really Web2.0, even though it was, and the usage felt off. It didn’t make intuitively as much sense as G2, though it was still far better than Coppermine (which frankly I hate, and I know more people who argue with it than anything). At first I thought it was because I was so used to G1 and G2, but then I realized that over the last 10 years, I’ve used so many different systems that I’m fine with subtle differences. I’m savvy, I’m smart, I can code, so why did G3 feel wrong to me?

    It was too hard. Too much was built in and not plugable. Too much was hard coded in itself. Theming was impossible in the first release, and way too hard in the third. Understanding the theme system in G2 was easy, though implementing it was hard. Understanding it in G3 was hard and implementing was horrific. And before someone reminds me, AGAIN, that this isn’t even a beta product but an alpha, quite frankly that’s not an excuse. The basic things you need to be able to do with a first public release (be it beta, alpha or whatever) is to use it: Upload photos, change options, theme. That’s it. Those are the three things at it’s most basic that photo gallery software has to have, or you may as well be using an off-site solution.

    And while I may sound like I’m ranting, I’m not. I’m sad and frustrated and … You know, I really like Gallery! I really do. But it was starting to feel like Movable Type. They made a big shift and suddenly I wanted to know who peed in my coffee. The code felt wrong, it felt klunky, it felt raw. It was like starting over, and I didn’t like where it was going. And I realized the fact was that I was going to say goodbye to an old friend.

    Personally I’m all about the simplest, best, tool for the job. I wanted a way to update news on JFO and, when that was ALL I needed, I used CuteNews. When I realized the site was going to need something more, I weighed my options, tested software, and decided that while WordPress was a bit of overkill, I knew how to support it and customize it to be what I needed. In the end, that proved to be a perfect choice. When I had a forum (the first time around), it was IPB, which I liked, but it always felt too big. Now I use the very basic bbPress and it’s what I need and nothing more.

    If WordPress had PhotoPress, I’d probably have snagged that. Instead, I shopped around. I installed Coppermine, again, to test. I put up G3-alpha3 and then 4. I went to WikiPedia and dug out the compares and ended up in a head to head battle between ZenPhoto and Gallery3.

    ZenPhoto won by feeling better.

    Seriously, it’s asthetics at this point. There are only two features I miss: Being able to re-upload a picture and keep it’s MetaData, and having ‘new’ images show up with a different background color. But I can live without those.

  • Why not homogenize?

    For another site I maintain, I use a total of six different coded products. Not one of the lot is actually integrated with the other, and no, I don’t use the same password between them. I’ve been having thoughts about merging the various tools into one vended support option, but as I look into the options, not one meets all the goals I have.

    Most of the time, when people look at a One Ring solution to keep all their products in line, they think of two aspects: usability and style.

    Usability means that, on the back end, you only have to learn one style of tools. We all know that no two product vendors produce the same style of code. Microsoft’s suite of word processing tools are, by far, the best out there, but Photoshop’s the place people go for photo editing. The interfaces between the two software tools is nothing alike. They’re so far un-alike it’s laughable. Hell, even Word on Windows is dissimilar to Word on a Mac.

    So for usability, people like things to look the same, or at least similar, so they don’t have to think hard when wanting to make a change. That makes sense.

    Style is more complicated. They want things to look the same. This makes more sense when you’re talking about a webpage, were you might have multiple background tools, but you want the whole site to look the same. This is called seemless integration. I call it style, since it’s a look and feel situation for the end use. Style points are useful. Style points keep people coming back to your site.

    In web software, which is as far as we’re going today, seamless integration is the thing. is content, a point most sites I’ve seen tend to miss. You have to have something worth reading, or people won’t read. Second? They have to enjoy the visit.

    A pox on all the sites with dark backgrounds and light fonts! That is not enjoyable! Most of us grow up with black text on white/light paper! We’re used to it, we like it, and our eyes have adjusted. Pander to us!

    There, was that too hard?

    Once you get a design, folding your multitude of tools into a seamless integrated design is fucking hell. Period. The majority of my sites are hand-coded, which means any integration was done manually. Over the years I got wise and used PHP includes, and then a PHP/SQL pastiche. But I attacked each part of the website problem as a separate entity.

    I needed polls, so I found good software. I needed an RSS feed, so I found one I liked. I needed a better gallery, so I picked on and so on and so forth. Doing things that way made extra work for me, this is true, but it also allowed me to tackle each new component as an individual. Would it have taken less time if I’d found an all in one solution? Yes, but it falls back on the problem that what I need doesn’t fall under usability and style.

    When I look for a new addition, I look for it as it’s own thing. A gallery needs to stand alone, without the rest of the site, and meet my organizational goals, my pretty URL goals, and my bandwidth goals. I’m confident enough in my l33t skilz to hack a system and make it look like how I want.

    So for me, a hacked up mishmosh system suite is what I need. Each tool is tailored specifically to my goals, and while it makes more work for me, the end user never has to deal with most of it.

    And if they’re happy, I’m happy.

  • Why Wiki?

    I like the site I have for my RPG a lot, but it’s very big and often unwieldy.

    So I started to think about the ways I could simplify it, make it easier to use and easier to navigate. The first idea I had was a Wiki. I want to point out that when I started with a CMS (which I still love for what it is), I spent a lot of days going ‘Oh, my head! I’ll never get this.’ So I expected the same here.

    What ended up happening, over the course of a month, was the realization that while the CMS is a little pain in the ass to navigate, it works better for what it needs to be. Everyone who uses it has a blog, and I’ve put so much effort into customizing it that really, the only part that would be helpful is if the ‘encyclopedia’ for the site was a Wiki. At that point, however, it wouldn’t be ‘linked’ with the CMS I have so I’m fucked. Not in a bad way, just in the ‘this works, it works 90% of the way I want, and there’s no reason to screw with it.’ If Slayer ‘reboots’ and we start a new game, I may try something different. CMS was an experiment, and it was successful.

    But … I do have another site that might benefit from an encyclopedia. About 150 individual pages of information, contextual and informational. Okay that was a little redundant, I’ve had a long week. But still 150 pages. And maybe, just maybe, updating it all, by hand, alone, is a bad idea. After all, there’s stuff I’ve never finished because I don’t have gobs of free time. And as proved by the forums on that page, people want to talk and help. So how can I combine that with an ‘easy’ to use back end, and a not-terrifying to use front end?

    Wiki.

    Okay, so it’s got a crappy admin side and if you’re not a coder you’re going to cry setting up MediaWiki, which I did and I am one. And some users are skeptical about how to create pages and … it’s been 2.5 months and of my 32 users (yes! 32!) four people (besides me) have added information. Hell, having one person add information is helpful to me, so I call it a success.

    There are 1314 total pages in the database. This includes pages about the Wiki, minimal “stub” pages, redirects, and others that probably don’t qualify as content pages. Excluding those, there are 291 pages that are probably legitimate content pages. Of the 291 pages, 150 or so came over as a direct copy/paste import (and some clean up), and 140 or so were added special to the wiki because I had more ‘flexible’ room.

    That flexibility is what drove me to Wiki. You know, when you make a new website, it’s a pain in the ass to link everything up the way you want it, and you have to come up with some sort of structure that will make sense to everyone and you hope they can follow it?

    Wikis piss that out the window and laugh at your ancestors.

    That’s a joke, son.

    Wikis aren’t linear. Wikis are fluid and organic. They grow in the direction they grow because there is information to grow in that path. Not to say there’s isn’t a rhyme and reason to the site, but a Wiki accepts the fact that things criss-cross and double back and take weirdo curves and twists when you’re not looking. Of my 291 pages, only 61 don’t link back to something else, and even so, I’m taking the time to go through those pages and cross-reference. Yes! That pain in the ass, time consuming project of ‘Doesn’t X refer to something over on Z?’ A Wiki links all those up with a simple, easy code [[Page Name]]. Thats it.

    Okay, if you’re an HTML coder (guilty), your brain has a moment of pain looking at this. Another fsking pseudo code language to learn? First HTML, then BB code and now Wiki Code? 90% of the HTML I’d want to use work on a Wiki. The only ‘argh!’ moment I have is with headers, and even then it was easy to fall into that code mind-frame.

    So what did I learn?

    To run a Wiki you have to let go of control. And that’s really hard. If you have a ‘normal’, let’s say traditional, website, you have a small number of people who can update the site. You have a set design the site maintains. You have templates and standards and such. For a Wiki, you have to step back and say ‘This is the site, this is how it looks. Please add more information, but we’d like to keep the feel we have.’ Yeah, you can follow along behind people and clean up what they do (some people use wacky grammar, others use l33t speak), but essentially you’re giving room for their voice.

    You can’t take that away once given.

    Well, you can, but you’d be an asshole.

    Running the Wiki has made me a ‘kinder’ person, sort of. I’m more laid back with people who post in perplexing grammar on the forums. I’m not going to be a firm rule-mistress. Okay, maybe on the Wiki I’ll be a little more the enforcer, but there’s no reason not to give people chances.

    Of course, I still can’t figure out what the hell “I love to but deberia to allow him grissom sara to love it also” means!

    By the way, why did I choose MediaWiki, over something more use friendly like TikiWiki or something ‘prettier’ or something easier to hack? Well, I actually tried out a handful of Wiki’s before I settled on this one. I’d load them, make a couple pages, toggle around and see what I thought. The one I picked was the one that felt right to me. You might think differently.

  • CMS: What is it and why do I care?

    CMS. Content Content Management System.

    A website where the admin, as a user or as a group of users, are able to control the content.

    If you’re anything like me, you read that and said ‘Yeah, and your point is?’ I’ve been poking around CMS style websites for the better part of a year, and it wasn’t until last week that I dipped more than my toe into them. Part of that had to do with the age old square peg/round hole syndrome. The rest was fear. But I’m here to tell you that CMS isn’t anything to fear! In fact, I’ve decided that for one specific instance, it’s perfect.

    The final decision, I blame on my web-host and my office. They sent me an email about a new tool for my website. Fantastico!, a tool that can install ‘programs’ to your website. By program, they mean blogs, portals/CMS, support tools (for your users), forums, eCommerce tools, image galleries, and ‘other scripts.’ I laughed a little, because I’d already managed to install a blog (MT, which isn’t offered by Fantastic), forums and a gallery. I had already mocked up a static style website but as I was about to click away, I saw the list for these portals/CMS thingies.

    My head spun and started to hurt a little but right away I recognized PHP-Nuke and Post-Nuke. Now, you should know that to look at those web pages is painful. They’re NOT formatted for people like me. They’re a cluster-fuck of information, scattered in an organizational mishmash that makes your desk look neat and tidy. Yes, they’re that bad.

    Why in the name of G-d would I want to use it? That’s best answered by showing you two examples of websites.

    Example One: The Traditional, Static, Informational Website

    There’s only one main admin to this site, me. I take the burden of collecting articles, images, summaries and any related information posted anywhere on the net and compiling it into something that makes sense. The site’s been up since 1997, and there are around 400 individual static SHTML pages to the site. I use a blog to keep track of updates, but for the most part, when I add a new page, I do it manually. Sometimes other people email me things I’ve missed, and I update that as well. It’s a once-a-week updated site, with a mass email that goes out every Thursday.

    Example Two: The Collaborative Website

    There’s still only one admin to this site, but the information comes from up to ten people, all of whom keep ‘journals’ of their personal activities and post public ‘news’ about the site. While I still retain the right of final inclusion on the site (should it stay or should it go is all my call!), most of the time I’m just going to tweak a format change. This means, with 10 people, I could get up to 50 request for new information a day. One option was to run a blog, where there are static links on the site for the ‘general’ information, and then a blog for each person and one for the group as a whole. The other was to have a 100%, updated by me and me alone, static page where they email in their new data and I upload it.

    Yeah, Example Two is starting to may your head hurt, isn’t it.

    I tried Drupal, which installed fast and had the best documentation, didn’t give me enough flexibility. Geeklog didn’t install well and the documentation was very geeky. PHP-Nuke has no documentation, and Fantastico! goobered the install a little. phpWebSite, for some reason, didn’t work at all, though it looked very nice. Post-Nuke was similar to PHP-Nuke. Siteframe had one idiot error I couldn’t solve. Xoops … well, there was no reason for me to try it, because I went back to PHP-Nuke.

    Why PHP-Nuke if it had no documentation of value and the install was goobered? Because when I ran a Google search on my error, I found the solution on the second hit. Really, that was it. Once I had the meat of the site up, I figured I’d better try it out.

    My head hurts today, thinking about the hoops and ladders I jumped around getting it to look right. Step one, I decided, was figuring out what I wanted from the site.

    1. To be able to have my users log in and post information.
    2. To be able to organize that information by Subject.
    3. To have an archive that worked by Subject.
    4. To have individual journals so each user could share their own information

    Doesn’t sound too hard, and all that was included in the basic PHP-Nuke Package. On to Step two. Make it pretty. PHP-Nuke uses a concept called ‘Themes.’ Each theme controls the basic layout of the page, the colors and the images. After staring at one for a while and making tiny changes, I figured it out and was done. Not too shabby. The problems I had with the layout was not the theme, in the end, but the ‘Modules’ and ‘Blocks.’ Unlike Themes, the concept of Modules and Blocks was less idiot-friendly. Blocks are like Building Blocks; you place a block in one of three positions ‘Left,’ ‘Right,’ and ‘Main.’ Okay, not too bad. You place ‘Modules’ inside blocks, and you can place multiple if you want. I wanted everything that was a menu block to be on the left side. Done. Now ‘Modules’ themselves were painful.

    Modules are all the special frippy website tools. The FAQ is a module. The Categories are a module. The News is a module. The Topics are a module. Once I understood that, I had to ask what the difference was between a Category and a Topic. In the end, I wasn’t really sure. Ipstenit (my test subject) bitched that in her head, they’re synonyms. I happen to agree, but as I argued ‘It’s not my motherfucking terminology, help me rename them!’ She wondered why the fuck I’d want to use a craptastic system like that.

    The benefits. Once I could get it up, I was sure it’d be pretty and useful! I think it is. At least, Ipstenit was able to use it without too much hand holding, and most of the problems we solved by copious documentation.

    Anyway, my Topic/Category solution was such: Topics are ‘Subtopics’ and Categories are ‘Main Topics’. Our current topics are ‘Actors,’ ‘Characters,’ and ‘Episodes.’ The Categories are ‘Cast (Season One)’ and ‘Episodes (Season One).’ If you click on a link for ‘Cast (Season One)’ you get a list of all cast (characters and actors) for the season. If you click on the link for ‘Actors,’ all you see are the last 10 Actor posts. It’s not perfect, but it works. What we really need is Subcategories, but those don’t exist code-wise yet.

    So in the end, what’s my advice?

    DO NOT install any Portal/CMS software unless you’re a coder or at least very comfortable with SQL and PHP. And if you are a coder, have the least code-friendly person you know test it and try not to get irritated when they ask you questions. It’s freakishly complicated stuff, mostly poorly documented, and if you’re really new at all this, you’re going to tear your hair out.

    90% of my problems came in making it something my non-techie users could understand. Once I broke that barrier, the 10% left were cosmetic. In that 90% was a lot more coding than I should have had to do, in my opinion. The archival options were for month only (January 2004, February 2004 etc) and I found I wanted more than it offered. On the flip side, setting up a theme was as easy as designing a web-page and uploading the required files. Also, adding in new modules was a snap! It’s a lot of give and take.

    IF you need a lot of people to be able to add information to your website, it can be very useful. By no means is it a catch all solution, and even now I’m wondering if I could have found a better one. It’s been a good learning experience, and there’s a lot more about coding that I now understand. It’s not for the timid or the code-shy, that’s for fucking sure.