@ewolff Of course not :-) But the best way to make a bad architecture work is to improve it.
1
1
@stevevinoski @pavlobaron Even there, I'd claim an Akka-based system can be much more similar to an Erlang one than two Java-based ones
1
1
@stevevinoski @gregyoung Which is exactly my point :-) The language doesn't matter that much (unless there's just one for that platform)
1
@sascha_vogt @ewolff I'd have to disagree with that. Bad architecture is almost by definition the one thing you can't get over
1
CollegeHumor nails it: “Why Can't You Use Phones on Planes?” collegehumor.com/video/69057…
2
28
11
@sascha_vogt @ewolff I was only considering runtime aspects. If developer productivity is included, languages matter a whole lot.
Architecture matters the most. Runtime platform matters a lot. Language? Not as much as you think.
7
48
13
@dierken @sebastiangozin @ctford You are right; it's more like "see it work first, then fail, then change".
1
finished Into the Black: Odyssey One by Evan Currie and gave it 3 stars. Solid stuff. amzn.to/HYLYWX
1
Spent an extremely enjoyable week in Jersey (and even ran into @simonbrown in St. Helier). What a great place.
1
1
@sebastiangozin @ctford Interestingly, it seems everyone just has to learn this the hard way – by seeing the easy solutions fail first
2
@tomayac @manusporny @danbri People I talk to (sadly) only ever care about Google. So an RDFa Lite endorsement would clearly do wonders.
1
@tomayac Understood. Still, @manusporny has a key point in that schema.org should be open to outside contributions /cc @danbri
1
@mfhepp @raymcdermott What are Microdata's advantages over RDFa Lite?
2
@SeanTAllen Mostly: changing the architecture, to a lesser degree: changing the runtime platform.
2
1
If you believe Twitter's scaling story is about moving away from Ruby, I think you're really missing the point
1
7
3
@kellabyte I fully agree. Really smart move to not keep this secret IMO.
1
“New Tweets per second record, and how!” blog.twitter.com/2013/new-tw…
1
10
7