Thursday, June 24, 2010

Tip On Web Service Engineering

On my last project, I'm sending data back to a client. The client is taking the data and updating several internal systems before acknowledging the data is received by sending me back a success code.

This is causing me all sorts of head aches. Their internal systems a slow, and I'm getting some strange errors. I upped the timeout to 10 minutes even, and still no love.

So, some advice. Just pass data to a store. Have your clients and yourself do their processing on that data outside of the web service. This will speed things up quiet a bit and help narrow down issues, and avoid timeouts ;)

If need by, you can setup your status's something like this.

  • Data Sent but Save Failed.
  • Data Sent Successfully.
  • Data Processing Failure.
  • Data Process Success.

Thursday, June 10, 2010

Linden Labs on the Decline

Linden Labs, the makers of the virtual world Second Life, announced a 30% layoff.

That is a shame. To me that kind of spells the end, though it will be a slow one, and Second Life will never completely disappear, as it is open sourced now and has a solid core group that love Second Life.

My experience with Second Life goes back when I walked into a co-workers cube and saw his Second Life avatar flying over a virtual landscape. Crappy graphics, things didn't rez in very well, he and I basically po-poed it. But as I left his cube I saw something, and I hate to say it, something that caught my attention. I'll get to that in a minute.

Then I heard about someone selling virtual land in Second Life for big bucks. Really? I thought. Our V.P. at the time seemed to thing this Second Life thing might be of use for educational purposes, and my co-worker who originally explored Second Life didn't seem much interested in pursing it, so I thought I would take a peak.

Well, long story short. Enthralled by Second Life's potential, was like viewing the future. Built a virtual university with others for demoing. Went no where. Second Life was too bugy, lacked a few key features. Met my future wife during the process though. Ended up getting a new job and leaving Second Life behind me.

I've got a lot to right here, but now isn't appropriate, so I'll come back. The short is, Linden showed us what could have been, but couldn't deliver, partially because the technology (bandwidth mainly), partially by design flaws built deep into the platform that were too costly to fix, and partly due to not really understanding what they had, a real "Second Life" isn't here yet.

Props to Linden Labs (Corey and Phil, plus a few other unknown to me engineers) for taking a great stab at it. I wouldn't consider the upcoming failure of Second Life a persona failure. You guys laid the ground work that probably won't be picked up on as soon as people think, but one day will. Thanks.