Profile cover photo
Profile photo
Waldemar Kornewald
399 followers
399 followers
About
Waldemar's posts

Post has shared content
Now this is pretty significant for web developers.
I've been meaning to do this screencast for weeks, ever since I first showed the screenshot of SuperDevMode and SourceMaps working. +Brian Slesinsky has been working non-stop polishing up SuperDevMode and now it's a joy to use, I figured it was time to give you small demo in real time of how SourceMaps work in Chrome. Edit: I forgot to mention that Firefox will also support SourceMaps.

Post has attachment
We've published a video about our wireframing software and are asking for feedback on Hacker News. Please join the discussion.

Post has attachment
Amazon DynamoDB is a new hosted NoSQL service. It's replicated across at least 3 data centers, scales automatically, and has low latency. Really nice.

But the query capabilities are very limited (comparable to simple key-value stores without secondary indexes). That means extra work on your side to emulate secondary indexes - which you'd get for free with other NoSQL solutions. I wish they'd at least provide manually specified secondary indexes (similar to App Engine), so they don't need to index everything, but still make development very easy.

Also, I'm not sure if the costs are acceptable because the pricing is difficult to compare to other services and hosting on your own servers. For example, is $1/GB acceptable? App Engine's datastore is also replicated across at least 3 data centers, so how does DynamoDB's $1/GB compare to App Engine's $0.24/GB? Is the latency so much lower that it causes the 4x higher cost? Does App Engine have a higher overhead, thus causing much higher costs than the $0.24/GB might suggest?

At least, DynamoDB looks like a more attractive alternative to SimpleDB. Complexity-wise it makes scaling much easier, but queries/indexes much more difficult.

SimpleDB vs DynamoDB vs host some other NoSQL DB yourself? Pick the lesser evil. However, if we had DynamoDB with secondary indexes that would be really really interesting.

Post has attachment

Post has attachment
Yay. App Engine 1.6.1 will introduce three different frontend instance classes, so you can have more RAM and CPU performance. Really nice.

You can now also play with the experimental file conversion API (e.g. .doc => .pdf) which even features OCR for images! How cool is that!?

And finally the dev_appserver gets a nice speed bump. Still, it's single-threaded which causes headaches with multiple HTTP connections blocking each other. Let's hope they'll make it multi-threaded real soon now.

Post has attachment
Nice talk about inbound marketing:
http://hackersandfounders.tv/RDmt/rand-fishkin-inbound-marketing-for-startups/

And a short summary comment of an important question on Hacker News:
http://news.ycombinator.com/item?id=3303046

Post has attachment
Finally the PyPy team is playing with solutions for reducing JIT warmup time by saving the JIT state. Day-dreaming: Maybe we could enable aggressive/costly JIT optimizations, run our code, save the JIT state, and then reduce optimizations from "aggressive" to "normal" in the release version, so the JIT won't take up most of the processing time.

As always, there are lots of other cool optimizations. Click the link to see the complete update.

Post has attachment
Slightly frightening and awesome at the same time. :)

Post has attachment
App Engine SDK 1.6.0 released with dev_appserver support for Python 2.7

Post has shared content
The new App Engine billing model is now live:
http://www.google.com/enterprise/cloud/appengine/pricing.html

If you have a problem with your billing or quota please send an e-mail to:
appengine_updated_pricing@google.com

We are also monitoring:
http://groups.google.com/group/google-appengine
http://code.google.com/p/googleappengine/issues/list
and Stack Overflow
Wait while more posts are being loaded