Tumblelog by Soup.io
Newer posts are loading.
You are at the newest post.
Click here to check if anything new just came in.

July 22 2010

Ah, and about the recent momentary lapse of judgement

aka downtime: We are trying hard to reduce these and are building a new infrastructure. I know that this excuse doesn't buy anyone a lost Sunday afternoon, but we are really taking an effort juggling all the clubs that are currently in the air - some of which are burning, too. And we get better at the juggling thing ;)

But it's all good news, really: we have the resources and opportunity to do something wicked, and took the extra 20% stress points to make it truly awesome. Or so we hope.

Late night hacking @soup.

Just awesome. @c3o and @sethx are working to make some sweet sweet javascript magic happen (which will be an essential stepping stone of the next big soup goodness)  and are trying to reverse engineer the dojo toolkit. Or something like that. 

By the way, if you are into coffeescript: Please read http://gfxmonk.net/2010/07/04/defer-taming-asynchronous-javascript-with-coffeescript.html - and help gfxmonk making it happen. Or improve the concept by contributing. This would REALLY make me consider "javascript" as an option for my next skill-to.acquire.
Reposted byelpollodiablo elpollodiablo

July 05 2010

Goodbye, scribd.com

This is beyond me. We liked scribd, because it allowed us to display uploaded PDFs fast and appropriately. And surely, they liked getting new users.

And what did they do? They had their iframes escape, so that a Soup that displays a Scribd embed redirects to scribd.com as soon as the embed was loaded.

What did they expect to achieve with this? 

More users? Not. More annoyed users? Maybe. Less presence on platforms like soup? Surely.

June 30 2010

Long night hacking on the API

and I like it, I rrrrreally do. But now I have to untangle a huge knot in my head. Epic.

@meta: expect to hear from me soon :)

For all the others: We finally came back with post by email, which took us an embarrassingly long time to restore.

Wanna know why? Because we hit google mail's lnbox limit. Yes, you read right. It is easy to check with dnstools, soup's emails are routed to google. It is kind of a hack, but i like it. It is a good spam filter and the service is very reliable and cheap, and removes a huge deal of (recurring) workload.

But it has bandwith limits. Uh-oh.

Well, suffice it to say, we had other things to do (tm) the last few months^W weeks, and until the new imap server was set up and the import script was fixed ... well, it took us an embarrassingly time. We know. We are sorry. I feel with you, I really do - HTC is consequently delaying the Android 2.1 update for my hero - for more than 6 months now! 

On the plus side, John, our iPhone guy made some progress (as I heard), and Chris, who will be doing Android, had his first talk with UI designer in charge @c3o. We are immensly proud of our two mobile chaps.

Expect a lot from us in the next few months, and please forgive us some of the bumps... we had and have a load of overdue work and meta-crap to do.

June 17 2010

The long promised API

is in the works. The read part of it will be a fully fledged graph API that will enable developers to slice and dice all of the relations we can think of that exist between soup entities like users, posts, blogs etc.
The write part will probably be just some kind of http POST submission thing, enough to get started though.

In two weeks or so, we'd like to have some beta testers for the first version of the API, just to check if we are on the right track, get a feeling for the performance impact and, of course, to build up some initial developer relations.

We are a developer driven start-up and as soon as there is an API that allows us to give away access in an efficient and secure manner, we will let you know anyway. But it would be mighty interesting to include those of you who are really invested in their ideas in the process before the official release happens. We are talking not only Soup-only addons but also your own (probably already existing) webapp which you could feed content and users from soup. For bigger projects, we can even talk about financing your summer.

So, if you have ideas what you'd like to implement (or already implemented), please reply to this post or contact us at team AT soup DOT io. If you are in Vienna or could stay in Vienna for the summer, include that information too.

And before the holy war starts: there will certainly be no invisible or irresponsible invasion of your privacy like at facebook and friends.
Reposted byelpollodiablosobasebnereveryonecoloredgrayscalemondkroeteworm23frank5bong0ciacon

The friggin' missing posts and 500 errors

should be gone now. We now restart our memcached every night in the low-traffic-time. The effect of which is that we rebuild the frontend cache during the slow traffic buildup in the early morning - which should not be noticeable. Also, we now restart our rails servers on a regular basis, as so many other rails sites do, so the lot of 500 messages of the past weeks should stay in the past, too.

On a side note: we are working on fixing the underlying problems too as far as memcache is concerned. However, the memory problems of rails we are probably not gonna fix by changing our code - rather by trying out other versions of ruby enterprise edition.
Reposted byhairinmy hairinmy
Older posts are this way If this message doesn't go away, click anywhere on the page to continue loading posts.
Could not load more posts
Maybe Soup is currently being updated? I'll try again automatically in a few seconds...
Just a second, loading more posts...
You've reached the end.

Don't be the product, buy the product!