I’m happy to report the Zen Garden archives have (finally) undergone a major re-tooling.
The css Zen Garden has been desperate for some TLC for a while now. Submissions have been increasing steadily (sometimes 6 or 7 per day), but management and archiving has been a frustrating problem for quite some time. So I fixed some of the major problems.
Since its launch on this domain over two years ago, the css Zen Garden archives have undergone numerous changes to cope with the growing volume of designs.
It seems hard to believe now, but at one point in time all designs I had received fit on one page. When that page grew too weighty, I chopped it into a handful of categories and gave each of those their own page. Then they had to be further subdivided into their own series of pages.
However, the categories themselves have become mostly meaningless, with ‘Conceptual’ becoming a default placeholder for just about any non-official design, unless I found a way to cram it into one of the other categories instead.
But fixing the problem either meant way more work for me, or adding a large degree of automation to the publishing process. So I opted for automation.
After spending the better part of last year writing a book on the site, it became obvious that my manual addition process was going to need to be addressed, and soon. So in January or so I started tinkering with MySQL databases to see what I could come up with.
By February, all new submissions were going into a database. Not a very good database, mind you, but it was enough to allow me to write a quick manager for the site during the spring. All submissions went into a database, which allowed me to categorize and automatically generate HTML that I could copy and paste into various templates. Handy, but not good enough.
About a month back, I needed help screen-scraping the list of Zen Garden designs. This was to facilitate Version 2 of the manager, which would contain a full database of all Zen Garden designs, so that publishing would be automatic from now on. And barring some publishing settings which will take a bit longer to enable, it’s finally done. The publishing workflow is almost completely automated, so all I have to do upon categorizing a design is hit the big ‘Publish’ button. (Official designs take a bit more manual work, but that’s a necessary evil.)
So now that I can update the site without planning my schedule in advance, what can we expect?
Well immediately, you’ll see that the archives finally have screenshots. It’s been a long time coming, and the lack of having them has bugged me too. Now that I spend less time updating, I can spend a bit more time to grab these for the official designs. This is a manual process, which involves Safari and Photoshop — until it can be automated on the server as well (and I have my doubts this is possible, barring an API from BrowserCam or similar), screenshots will only exist for official designs. We’ll see what this does to my bandwidth though…
As well, I can do fun things like creating ‘recently added designs’ lists, and even offer an alternative archive format in the ‘View Designs By Date’ page. This is a bit broken for existing designs due to missing publish dates — creating metadata manually is no fun at all, thanks — but it ought to become a lot more useful as new designs are added.
The categorization problem above? So fixed. Now instead of a handful of meaningless categories, designs can exist in any number of the 50+ categories which describe things like layout type, dominant colours, and themes. I’m basically treating them as tags, and can add new ones at any time. For now, I’ve only re-categorized the 180 official designs, but as new ones come in I’ll be more accurately categorizing any and all that get published. The new problem is going to be categorizing the categories; I’ve reserved a ‘group’ column in the table for precisely that, whch will be implemented in due time.
What else? Oh, a couple new RSS feeds, which are going to be ‘experimental’ for now. The designs listed are screwy at the moment (expect them to be completely innaccurate) and I’m not 100% convinced they do what they’re supposed to yet, that will take a few updates to figure out. But you can expect they’ll eventually work.
So that solves a few problems, and leaves a couple of questions hanging about what has yet to be done:
- How about Comments/Ratings?
- I’ve considered comments, but I’m not going to do it. Browse any major CSS design listing site to experience first hand why not. Anonymous commentary-free ratings might be a possibility, but very low priority.
- No. Well, kind of. Since I’ve spent far more (billable) time than I should have on this upgrade, I’m likely going to run some ads on the archive pages for a few months. Only to recoup the costs, and then they’ll be removed. I have a specific dollar value in mind, so they’ll only run until that has been reached. If you have a cool product that’s relevant to the type of people likely to be viewing the Zen Garden, get in touch. Otherwise it’ll just be AdSense.
- Why are the archives still on mezzoblue.com?
Legacy and overhead. The site was originally launched on this domain, and since the purpose of the site is to re-skin a common body of content after all, picking just one of those and just using it for the archives and supporting material seemed a little silly. But forcing submitters to style more variations would have been more time-consuming for them, and the barrier to entry would be a lot higher. This was a real concern at the time, so I dropped the supporting material in this site’s template to side-step the problem.
At some point I think there will be a ‘master’ design for the supporting material, which will then see it and the archives pushed over to the csszengarden.com domain. At that point, the page that greets you when you visit the site will change to more of an explanation/design listing, to solve some of the lingering problems around the fact that I can’t change the markup structure these days. But I’m not quite ready for that step yet.
- Is it still relevant?
Truth be told, my idea of what the site is has changed quite a bit recently, perhaps yours has too. Keeping in mind that it’s a lot of things to a lot of people, and designers new to CSS are still stumbling across it for the first time every day, its original purpose hasn’t gone away. But for those of us who get CSS design now, and that’s a significant lot of us, the initial impact has long since faded.
What was reinforced to me as I went through the archives for screenshots and categorization, is what a wide and diverse body of work we’ve created. Hundreds of designers are solving the exact same problem in so many diverse and unique ways. That work continues to be relevant, and although much has changed about the techniques and coding we use in the past few years, a good design is still a good design. And there are a lot of them. And there will continue to be more of them.
As it says in the wording of the site, which is long obsolete, “One day this gallery will be a historical curiosity; that day is not today.” This may be true of the markup that makes up the site (more
classes than necessary today), the site navigation (woefully inadequate), and even the wording itself (oh how I wish I could edit), the main point of the site — the designs people submit — will continue to be relevant as long as I continue to receive them.