Ask a question
jxnblkjxnblk asked:
Is there some secret, magical trick to getting the new Tumblr audio player in your theme?

I’m glad you asked, because it took me forever to figure out why it wouldn’t work. I originally had the {AudioPlayer} and {AudioEmbed} tags both in my theme wrapped in their existence conditionals, but it always showed the old player. So I switched it to only the {AudioPlayer} tag because that tag makes the most sense, wording-wise, but that appears to be the old one. Ended up using {AudioEmbed-640} only, without the conditional, which works. You can use the other sizes (250, 400, 500) but you can’t customize the colors.

Ask a question
novemberalphasierraechonovemberalphasierraecho asked:
Your site designs look pretty cool. Have you considered releasing/selling them?

Every time I build a new template I think to myself, I should clean this up and release it. And then I get to the part where I need to support all the various niche Tumblr features, add things like detailed likes and reblog info, create customizable colors, etc, etc. So I give up. I’m also not a huge fan of other sites looking exactly like mine while I’m using the theme.

This most recent design is very simple, code-wise, and would probably make an easy transition, so perhaps when I move away from it I’ll package it up and release it. I appreciate the interest, though.

Ask a question
alexstonealexstone asked:
This isn't a complain by any means or a plea to get you to stick with Tumblr. Just looking to better understand Svbtle. I'm genuinely curious what brought you to subtle. Full disclosure: I work for Tumblr, but I'm fascinated with how people view various platforms.

Great question, and I’ve gotten various versions of this since I announced my Svbtle blog. Most people want to know why I am moving away from Tumblr, or what will happen to Maniacal Rage.

First, this site isn’t going anywhere, nor do I have plans to stop posting here. I’ve been publishing Maniacal Rage in some form since 1999, before I even owned a domain. Since 2001 it has been at maniacalrage.net and I’ve used my own PHP CMS, then a different PHP CMS, then Movable Type, then yet another PHP CMS I wrote (this time using PHP to generate XML with XSLT generating HTML, ugh, insanity), then my own Ruby on Rails CMS SimpleLog, after which I finally moved to Tumblr where I’ve been for the past few years. The only reason I mention all that CMS insanity is that I want to drive home the fact that Maniacal Rage has been a part of my life for more than a decade, and I have no plans to abandon it.

Over time, though, the site has changed into more of a filtered feed of my creative interests—I post my favorite photos, videos I make, (what I think are) funny conversations with friends, and short thoughts or jokes about technology. Recently, I’ve yearned to really write again—to write long, thoughtful posts about technological minutia I spend an unhealthy amount of time thinking about. I’ve felt for a while now that this site has moved away from this kind of content and my audience partially has as well. I started to think about a separate blog for just this kind of writing, and Dustin launched Svbtle at a very serendipitous moment.

I’m not leaving Tumblr, and I didn’t choose Svbtle entirely for the content management system, though it is nice for focusing on writing. I chose Svbtle because it offered not just being part of a network of bloggers I find interesting, but also a fresh start for my writing. When I browse the Tumblr Dashboard I feel overwhelmed by the amount of great content—as most people probably do—and it’s very hard to focus on longer entries people post. I still use an RSS reader for those kind of blogs, but I have a feeling most Tumblr users do not. So I often feel my longer posts get easily lost in followers’ Dashboards. That’s not a problem with Tumblr, it’s just the way people digest content here. Writing over at Svbtle, there is no Dashboard, no aggregate, no tiny version of my content that can easily be overlooked. It’s just free-form writing and you can focus on one post at a time.

I realize some people who follow me might be interested in both sites, which means I’ve given them yet another place to check if they want to read my content. This is why I’ll continue to post links here, at Maniacal Rage, to entries I write at Svbtle. This way, you don’t have to follow me here and go visit me there separately, you can simply click through to Svbtle entries if you’re interested. If not, you can continue to (hopefully) enjoy my shorter posts, photos, videos and links here.

I think Tumblr is great, and I truly appreciate all my followers here and people who support Maniacal Rage. I’m not planning on leaving you. Just think of Svbtle as my weekly column in some (possibly boring to you, but hopefully not) technological rants magazine.

Ask a question
jkspnjkspn asked:
Will you ever be releasing themes to the Theme Garden? Your old one and this one both look nice!

I struggle with this. I’d love to release some of my old themes, but the amount of time they’d require to clean up, make generic, and add features to is a little overwhelming. I am hoping to make my previous theme available at some point, probably as a very cheap premium theme.

Ask a question
swimmindustriesswimmindustries asked:
Really like the new look. I am just starting to move my site to Tumblr and wondering how you handle theme building. It's such a pain. I've messed around with TumblrThemr, but that has its own set of issues. If you have a non SUPER PAINFUL way to customize I'd love to hear it.

I actually find the theming system at Tumblr to be pretty decent. There are quirks, but once you get used to them, it’s pretty easy to get into a good workflow for creating a new theme. I tend to work at this sort of backward. I design the theme and HTML with real test content and then replace that with template tags when I’m ready to go live. Here’s my general path:

  1. Back up current theme HTML, CSS, images, as well as run a full Tumblr Backup (obviously, if this is your first theme/blog, this isn’t as important)
  2. Visit your site and grab the existing, rendered HTML. You’ll use this both as a reference of how Tumblr renders theme tags, but also to pull test content into your developing theme.
  3. Start with your new template by creating a new HTML page locally, using the content from your current, working Tumblr theme. I use TextMate and CSSEdit for local theme development, since CSSEdit has a great preview mode that allows you to work on CSS live and see the changes immediately. By the time you’re done, you should have a working version of the theme (albeit only one page of it) with solid HTML and CSS using the test content.
  4. Now you want to port this over to an actual theme by replacing the test content with Tumblr tags. This sounds difficult but it’s actually the easiest part. Use the theme guide as a reference.
  5. Make it live. You’re bound to run into little CSS issues and template issues, so this part is a little frustrating. It would be awesome if Tumblr had a way to test themes non-publicly, but they currently don’t. So make your new theme live and then just fix issues as you see them. Little tip: Make your CSS filenames unique each time you update them so users don’t continue to use cached versions—I use datestamps for mine (example: styles_091611_01.css).

After doing it once or twice, you’ll get the hang of it. Good luck!

New Design, You’d Better Like It!

Finally had a chance to put the finishing touches on a new design for this site. I’ve been thinking about this new design for about four months, and working on the actual code on and off for the past two weeks. The last time I redesigned was in April of 2010—nearly a year and a half ago. For me, that’s a personal record (I’ve redesigned this site 20 times in the past 6 years alone).

The focus this time around was:

  1. Simplicity. I wanted to remove all the clutter from the sidebar (and the sidebar itself), the footer, et cetera. I narrowed down all important actions and stuck them in one place. The links I felt were most useful will stick with you as you scroll.
  2. Photos. I take a lot of pictures and the default 500px width Tumblr uses is not great for showing them off. Luckily, their high resolution functionality and some CSS allows me to make them huge now. Love it.
  3. New text styling. I based this off iA Writer, which I love to death (and am using to write this post). The way iA Writer displays text with Markdown is terrific, so I mimicked some of that here. Bold text is a good example. Quotes, lists, headers—they’re all designed to look more like text in Markdown format.

A completely new thing for me: I based this design on a theme. I purchased the HIGH RES Theme by JSTN and used that as a base to customize and design on. I’ve been coveting the HIGH RES theme for months now and it was a lot of fun to make it my own.

In addition, I accidentally and shamelessly took Will’s idea for a rounded-avatar header. He claims he has forgiven me, but I’m not so sure.

Credit Where Credit is Due

People who read regularly know that over the past seven months or so I’ve been very critical of Tumblr, the service that has powered my site since 2007. Starting back in January I wrote a post titled “Time for a Change” wherein I lamented the downtime and feature bugs and announced my plans to move to a different service. Then, in March, after a period of several weeks of many outtages and lots of downtime, I snapped one morning and created a jokey-but-scathing alternative error page for Tumblr. I knew it was harsh, but I also knew that for several months my website was down all the time and it was taking its toll on me. That post received a total of 612 likes and reblogs, which told me I wasn’t the only person fed up.

This led, that same day, to me posting a followup explaining how badly I wanted Tumblr to charge for their service, and why I thought it would lead to a better, more stable platform. This post received nearly 300 likes and reblogs once again confirming I wasn’t alone.

And so it went. I continued to bitch and moan, and the service continued to suffer. After my critical posts, the entire team at Tumblr unfollowed my account (and those following me on Twitter also unfollowed there). David Karp used to like my posts regularly and give me nice comments and feedback occasionally. We exchanged emails here and there and we were on very friendly terms before all of my complaining. Jacob Bijani would occasionally @reply me on Twitter and like my posts on Tumblr. These guys are very nice and I respect all the work they do. So I felt terrible as I watched them systematically unfollow me, but at the same time I couldn’t get past the simple truth that the service they were providing was failing me on a daily basis.

“I’m not here to make friends” has crossed my mind many times throughout my career in this industry, because it’s the opposite of how I feel. I’m here to make friends. I’ve made friends with so many fantastic people over the past 10 years, and I can’t imagine my life without these folks. So when my actions directly correlate to good, smart people turning away and never looking back, it’s a hard pill to swallow (related to this is the group of amazing people at Facebook whom I’ve also offended in various ways across the years, but that’s for a different day).

I sent David several emails about the problems with Tumblr and he continued to express both sympathy and urgency and an overall “we’re working on fixing it ASAP” message, but it just didn’t feel like anything was happening. For months, uptime was a roller-coaster and features would work sometimes and not others. It was initially frustrating, then angering, then infuriating every time I saw that goofy error page. Reports of Tumblr hiring more engineers kept surfacing, but no real-world results were visible to me as a user.

Under all my complaining and joke-making was a real, genuine message: I love Tumblr and I want it to be better. I kept mentioning that I wanted to pay for the service, pay to make it better, pay to support it. I reiterated to everyone who asked why I liked the service so much to begin with that the people who were building it were doing fun, interesting things and that the likes/reblogs system was great for getting your content in front of more eyes. I continued to preface every complaint with “I love Tumblr and I want to keep using it, but…”. But eventually I ran out of energy and gave up.

So here we are in June, just three months later. And what’s the situation like now? I’m happy to report it’s much better. Downtime and outtages seem very rare these days—I’ve only seen the error page once in two months. And, just recently, Tumblr has begun adding new features and redesigning the UI again. The new Messages functionality they released last week was a huge change that I immediately appreciated (I was finally able to answer boatloads of questions privately), and the new Dashboard design is refined and svelte. The future looks very bright for Tumblr, and I’m glad I didn’t jump ship back when the skies were darkest. That doesn’t erase what happened, but the team is doing a hell of a job handling the load and getting new stuff out. Credit where credit is due.

I’m sad to have lost the potential friendships, but I’m glad Tumblr is getting better. Just last week, in an Adweek interview, David said:

You get error messages at Google, you get exceptions when accessing the Facebook API. At this scale there are way too many moving parts for it not to happen. Desktop computers still freeze. When people post a screenshot of an error page and we know that they hit refresh once and happened to get an error, that does frustrate me.

He probably wasn’t thinking of me specifically, but I’m in there. While he’s perfectly within his right to be frustrated by complaints about a service he has poured his heart and soul into—and while I understand how crushing it can feel being a software developer myself—the other side to this is that if you act like there isn’t a problem or you take too long solving it, users don’t feel like they can trust you. Tumblr is earning back that trust now, I’m just sorry for the collateral damage.