pul

> leaving you susceptible to parked car doors suddenly opening

Dutch driving instructors teach 'the Dutch reach'. You get into the habit of opening the car door with your left hand, so you'll automatically look for cyclists that may be passing by.

Yeah, I started out labeling them, but that turned out to be extremely tedious. Does anyone know of tools that make annotating gifs easy?

This. With the slight nuance that it works better if you don't think of it as queues, since that suggests messages being removed when processed.

The big win when introducing Kafka is that (apart from the schema) it completely decouples producers and consumers of messages. That in turn reduces the required coordination between independent teams within an organisation.

I wrote this article a couple of months ago and didn't expect it to appear on my news feed all of the sudden. Thanks for the love everyone. Any questions are welcome of course!

Introduced just weeks after Confluent introduced their similar service [1]. Has anyone done an in depth comparison yet?

[1] https://www.confluent.io/confluent-cloud/

I fail to see how this would reduce the incentive to collect data. If anything, wouldn't it require _more_ information on users to figure out whether or not showing an ad would be worthwhile?

Could you elaborate on each of those four points? How did you come to these conclusions?

It ultimately comes down to trust. Some people lost trust in GitHub because of the acquisition. With Microsoft still aggressively pushing IE and Bing through their products, I can see why.

Now that Microsoft is moving from selling products to selling cloud services, developer trust is becoming one of its main currencies. Microsoft is trying hard to regain that trust.

It's still in Googles cache: https://webcache.googleusercontent.com/search?q=cache:zZwHCS... Not in archive.org though.

I've found that any math subject makes a lot more sense when coding it out in a language you're comfortable with. Especially when you write unit tests for it and think of edge cases. It forces you to reproduce what you've learned and reason about it.

Here's the link to what this article is referring to: https://worldview.earthdata.nasa.gov/

Not just that. Defining a successful startup as a startup that exited is a bit narrow of a definition. A bootstrapped startup that has sustainable growth and can remain independent is successful in my book.

Live demo at https://sbd.danilop.net/

Not fully, but implicit type conversion must be marked explicitly: https://medium.com/bigdatarepublic/key-takeaways-from-the-sc...

This is useful, good job.

Note that your search suggestion "Scala" also returns results with no reference to the language, but containing "scalable" in the text.

They're referencing the Bible: https://www.biblegateway.com/passage/?search=Ecclesiastes+3&...

> However, we discovered after some time that the custom Python implementation for those workers was dropping up to 5% of the events. This was mostly due to the nature of how reading happens with Kinesis: every stream has multiple shards (ours up to 50!) and each reading client would use a so-called shard iterator to keep track of where it was reading last. Since the used machines could always crash, be recycled, or scaled down, we needed to save those shard iterators in some serialized format to Redis and share them across machines and process boundaries. Since we had so many shards, every once in awhile we would skip events and hence lose them.

I've never worked with Kinesis, but in Kafka you'd store offsets specifically to solve this issue. When one of the members of a consumer group would drop out, the partition (read: shard) would automatically be reassigned to another member. This gives an at least once delivery guarantee, combined with idempotent actions gives effectively once semantics. No need to loose any messages. What was the issue that the dubsmash engineers were solving here?

Replace `least amount of code` with `least amount of complexity` and I completely agree!

Paper: https://link.springer.com/chapter/10.1007/978-3-319-60876-1_...

Blog: http://blog.trendmicro.com/trendlabs-security-intelligence/c...

Nice work on this site! I think I'll have something to do the coming time ;) How did you manage to find and index all these videos?

Scala 3 will include emuns as a language construct [1], but that will probably not be released before 2019.

[1] http://www.bigdatarepublic.nl/key-takeaways-from-the-scala-d...

Worst of all, only 8 of the 3306 links use https. 11 years really is an eternity in internet years.

Making it unprofitable would be the best defense against for-profit ddos attacks IMO. If using a botnet to mine regular coins yields more than ddoscoins the insentive is gone.

Can anyone think of defences not listed in the paper?

IT has now also executed ajax:

https://www.google.nl/search?q=site%3Adoesgoogleexecutejavas...

You might want to be careful with intentionally publicizing encrypted data, since it might only be secure for a limited time. Computation becomes cheaper over time (making brute force more feasible) and new vulnerabilities in the encryption algorithm can be found.

Going a bit off topic here: would a chrome/firefox extension that adds alt tags to images using AI be useful for you?

There are quite some api's that guess tags and descriptions for images. This doesn't look very complicated and might make inaccessible websites a bit more usable for you.