SVN Deployment

Seems simple in that “WTF is up with me not doing that until now” kind of way but setting up all of your web content in SVN and then using svn export to deploy to your production server just plain rocks.

  1. Version control–which is desirable on its face
  2. The model enforces backing up–your deployment is just a copy of your versioned repository
  3. Instant recovery
  4. Work locally, publish remotely–no zOMGWTFBBQ moments when your internet connection breaks the I/O pipe your text editor is relying on or even dropping your terminal session, whereby you lose your work

There are many things I’ve learned working in CM these last few years, but the value of SVN or version control generally, continues to be the lesson that keeps paying dividends.

If I were still rabbiting away at the computer lab at $SATELLITE_CAMPUS library, I’d be advocating to teach a mini course on version control as a tool for managing all research projects/papers. It is the hammer by which I identify and act upon all nails. 😉

Hi honey, I’m home

Looks like the migration is mostly complete. A little less painful than anticipated all things considered. Not a lot of planning up front and it looks like the main pieces came across with only minor hair pulling. I can’t believe 10 years of accumulated clutter on $OLD_WEB_HOST was this easy to move. Even if things are burnt, it’s too late to go back now. On my way out the door at $OLD_WEB_HOST I engaged in some pretty fun rm -rf behavior. Not often you get to pull those kinds of triggers.

So $NEW_WEB_HOST seems to be working out so far. Getting to a terminal window isn’t always all that fun and their auto-generated WebDAV thinger doesn’t work as advertised. Not sure I like the way $MAIN_DOMAIN (not blackfez.com) can be used to hit all of my “add-on” domains. Then again, one could do the same thing at the old place, but one had to use the company domain to do that.

I’m a tiny fish in a huge number of users now. CPanel is a nice upgrade from old host’s d-i-y interface. I don’t like how Python is a marginal language at new host and still no pre-fab Java servlet container. I do like that there is a support staff of >1 and that they have an always-on phone number to boot.

Long evening of keypunching makes it feel like work on the weekend so it’s looking like time to turn in. Let me know if you see something awry/broken. I’ll see what I can do to fix that up.

RSS as Memex

I know it’s a topic I come back to now and again, but only because it is one of the most useful “technologies” to come out of my WorldWideWeb experience. The ability to essentially bookmark something of interest and then have that thing tell you when something happens to it is a remarkable way to increase the number of things one can juggle. Layering audio and video files on top of the RSS framework gives us pod/vid casting. This has also reshaped the way I consume media to a huge extent.

Not having to search out content, but to let it tell you when there is something to consume frees up more time for consumption. I used to read blogs by clicking through all of my bookmarks, reading to remember where I last left off, and then reading onward until I’d consumed all the new postings. I could waste hours a day doing this and not really gain a whole lot of new information. I don’t think I’ve clicked on a blog bookmark in years now. I just follow links from my RSS reader.

Podcasting has made my life as a commuter so much more enjoyable. I currently timeshift about 15 radio shows from four continents on a variety of subjects. I no longer chafe under the constraints of what limited variety the local over-the-air radio provides. Podcasting also provides a platform for niche programming. You’d never get a weekly half-hour program featuring mashups that pull from classical, martial, and pop music on any corporate run radio station. I have two of these in my queue and when the mood strikes on a given drive, I can groove away.

Historically I’ve always subscribed to feeds for the long haul. A website I enjoy has a feed? Subscribe! I never did much pruning because I liked what they had to offer over time. There are some that changed in tone (or maybe I’ve changed a bit too) and I’ve deleted them. Some blogs are no longer active, much to my great disappointment. The subscribe/delete process tended to keep things at a pretty even level with my ability to keep up.

Lately, as in maybe the past six months or so, I’ve been using RSS a bit differently. Google provides a number of interesting ways to combine its search algorithms with the technology and it is easy enough to set up a feed and get news of a particular subject without having to go looking for it. For example, when Google comes across news matching the filter “Council Bluffs levee” I get a notice and link to the article. I’m following three general news stories this way. Google + RSS has become my own private clipping service.

Most online newspapers and discussion forums now allow you to subscribe to feeds on comment threads for news stories. This lets me track a discussion on the latest outrage to hit Metafilter, chuckle over the frothing masses huddled over the latest local crime sensation, keep up on discussions about the various ham radios I own, etc etc etc.

All of this flows through my email/RSS reader. The Web is clamoring at my doorstep, providing endless opportunities to wallow in a glut of signal (and noise) surrounding something that interests me. It is Vannevar Bush’s Memex realized. Using my reader’s indexing and searching functions, I can retrieve details and discussion history at levels that continually amaze.

The problem, and there’s always a problem, is that my reader/email client is becoming unwieldy. There are so many feeds I’ve subscribed to in the past half year that no longer bear fruit and should be pruned. The size on my filesystem of this vast amount of text and audio is starting to strain my ability to capture and properly manage backups for. I dread migration my primary platform so much that I have a hard time envisioning not using my current laptop for the rest of my life.

Still, there’s no way I’m going back. You can pry my collection of feeds from my cold, dead fingers. Something so simple as a standardized, machine-readable, XML format has enabled us to unleash the scouring power of software agents on the monstrosity that is the WorldWideWeb. This is the kind of thing that gave me shivers when playing with the building blocks in grad school. It’s almost AI, but it’s most like Memex…and we’re really only beginning here.

Looking forward, RSS-like applications are being bundled into the next generation of browsers. Visit a web site with a feed and the browser can remember the state of the site when you visit. Come back and it will jump to the place where you last were, or maybe even filter out the old and only show you the new, or maybe just highlight new content. Or maybe do all three and allow you the option to pick how you want it to behave. The money-making future of the Web is in aggregation + digestion + presentation + personalization + connecting similar datasets/profiles/people. Do this seamlessly and people will pay–either those who want this level of service or those who want to harvest the audiences these silos present.

Late night build support…

While composing a detailed “you broke the build and here’s how, but you’re not here so you’ll see this in the morning and have to answer the gnashing and wailing from the angry horde of testers” email, I was chatted up by hotandhorny13.

a picture of the chat window...you're really not missing much

I did not blur the URL to the probable malware site linked. If you’re fatheaded enough to type it into your address bar and browse that site, you deserve what you get. Just sayin’…

Finally!

A comment spammer left this comment on an earlier blog post

Finally, an issue that I am passionate about. I have looked for information of this caliber for the last several hours. Your site is greatly appreciated.

Glad I could procure something for him/her to finally be passionate about.

Wait, wait, don’t kill it

For what seems like forever, a certain list-serv list on a topic of academic research interest has been a mere dribble of traffic in my Inbox. Most of the traffic being the monthly “you’re subscribed to List-FOO” and occasional position postings, most of which are in the UK. I know I stopped monitoring the list in detail ever since I changed gears to a more dev-related role, professionally.

It would seem that someone on list had the temerity to post whether there was a discussion forum that was more actively addressing this academic question to the more-or-less dormant list. This being the academic silly season, a significant number of members are all twitterpated over a discussion as to why the list is dormant, whether the list is worth saving, and what can be done to rejuvenate the community.

Two interesting trends in the last two weeks:

  1. Long-standing personality conflicts are being aired for the first time in years
  2. Actual research questions are being asked again.

As to the second, I wonder where these people were asking their questions before the revival or if this is just guilt-based busy work, or maybe just a way to keep fingers in the pot. I also wonder just how busy the traffic would be if the initial question had been raised at a time that wasn’t between semesters. At any rate, it’s kind of amusing.