justinsb

58 ms·

1. 511.▲
   by justinsb 13y ago

   A great read. Another example that is near and dear to my heart... the CAP theorem (in the mathematical sense) is proved incredibly narrowly, and I believe there are plenty of ways around it.
2. 512.▲
   by justinsb 13y ago

   Dual gigabit, wi-fi and the ability to run Ubuntu makes this very interesting for a "home router / hub" type device. Video output makes it interesting for a media player device. Putting both in one low-powered box is awesom
3. 513.▲
   by justinsb 13y ago

   I think the main goal is not to make cloud software the commodity, but rather to make IaaS a commodity, so that there will be multiple vendors of IaaS that are largely interchangeable. There are today a number of public clouds that run Open
4. 514.▲
   by justinsb 13y ago

   I've tried Incero, Softlayer, Hetzner & OVH. I'd recommend Hetzner & OVH, and Softlayer if budget is less of an issue.
5. 515.▲
   by justinsb 13y ago

   "If Amazon faced a subpoena that required it to keep the order secret, such encryption would be useful to customers. “If the data is encrypted, all we’d be handing over would be the cypher text,” he said." That strikes me as a lit
6. 516.▲
   by justinsb 13y ago

   I think that "limit" is just FUD - can anyone explain where it comes from? With persistent SSH connections, the overhead of SSH encryption should be comparable to other encryption methods.
7. 517.▲
   by justinsb 13y ago

   Thanks - a nice explanation of Spark's benefits. The programming & execution models sound like big improvements - I'll check it out!
8. 518.▲
   by justinsb 13y ago

   I did find this (now-fixed) bug/enhancement: https://issues.apache.org/jira/browse/HDFS-2246 Sounds like, with that configuration applied, the in-memory performance difference between Hadoop and Spark should
9. 519.▲
   by justinsb 13y ago

   Can someone explain why the in-memory caching is such a big win? Does Hadoop MapReduce not do caching as well? I'd expect at least filesystem caching when the computation is running on the same machine as the data block...
10. 520.▲
    by justinsb 14y ago

    I backed this project. This is exactly the sort of blue-sky experimentation I think crowd-funding can enable, that otherwise likely wouldn't be economically feasible and/or fundable. I think I backed it as much for its ambition and vision
11. 521.▲
    by justinsb 14y ago

    I think there's a fallacy in there; just because something hasn't been done doesn't make it a bad idea. Otherwise we should just stay shivering in the metaphorical cave. But I think it's commonly accepted that there's a huge performance di
12. 522.▲
    by justinsb 14y ago

    Definitely agree - I don't think this aspect of 10x's focus will be easy. But that's what I'd be paying for as a 10x customer - that they've done the difficult job of telling people that are perfectly competent "thanks, but no thanks".
13. 523.▲
    by justinsb 14y ago

    I really like the emphasis on quality of talent. In my experience, too many recruitment agencies (the incumbents) are heavily focused on volume and optimize around their own compensation structure. If 10x can stay focused only on genuine t
14. 524.▲
    by justinsb 14y ago

    I think both experiments are consistent with the idea of thinking of the different Google properties as individual apps, and with Chrome apps becoming first-class apps. Thus on ChromeOS you would access gmail through the menu of application
15. 525.▲
    by justinsb 14y ago

    The AWS strategy is to release the bare minimum and then announce little fixes as if they were new products. It's gotten to the point where they release a product in one region, and then later announce availability in other regions with al
16. 526.▲
    by justinsb 14y ago

    "Democracy is the worst form of government except all those other forms that have been tried from time to time." Relational DBs & SQL aren't perfect, but they've proven better than 30 years of attempts to replace them. I think the burd
17. 527.▲
    by justinsb 14y ago

    Surely there is a per-site cache on the origin server (in what you call the "Listener")?
18. 528.▲
    by justinsb 14y ago

    Presuming this is RFC 3229, this is transport compression, not webserver offload. The response is generated by the origin webserver as normal. But rather than sending that response using the normal HTTP encoding, instead the proxy first do
19. 529.▲
    by justinsb 14y ago

    I think this is just RFC 3229, with a binary protocol (?) http://www.ietf.org/rfc/rfc3229.txt I've always thought there were some potential attacks there around cache disclosure (which Google avoided by going with SDCH instead). CloudFla
20. 530.▲
    by justinsb 14y ago

    Thanks!
21. 531.▲
    by justinsb 14y ago

    Will the "making of" videos be available to purchasers? That would be an incredible resource for anyone wanting to learn to make games...
22. 532.▲
    by justinsb 14y ago

    Anyone know if the plan is to fix up WINE where it doesn't work, or is the plan to write a Linux abstraction layer for their tools? Either way is good, but fixing up WINE would have benefits even for non-gamers (such as myself).
23. 533.▲
    by justinsb 14y ago

    It's normally considered bad form to catch NPE, particularly if a simple (x == null) check is the alternative. But that's largely a code-style issue; each to their own. I'm not sure what you mean by your static typing comment; I don't see
24. 534.▲
    by justinsb 14y ago

    My preferred approach here is to have two methods; one that throws and one that also takes a default value. If the caller would handle the not-present case by assuming a value then use the default-value version; if it's an error then use th
25. 535.▲
    by justinsb 14y ago

    Google provides Optional<T> as part of the Guava library, so this is a fairly well-known idea. I find it fairly tedious to use though. The @NotNull annotation seems like it should be the answer, but I don't know of anyone that uses
26. 536.▲
    by justinsb 14y ago

    I think/hope California is simply trying to raise awareness of the law, rather than believing that Delta's mobile app is a serious threat to privacy. On Android, the privacy policy is linked from the Play Store, and the standard Android per
27. 537.▲
    by justinsb 14y ago

    Do you work for AWS as well? If so, I'd ask that team AWS spend less time astro-turfing on HN, and more time documenting your systems, so we can assess these risks for ourselves. For example, I haven't heard of any precautions taken agains
28. 538.▲
    by justinsb 14y ago

    That FAQ you yourself pointed to doesn't mention that there are common components. We now know that when Amazon said - in that very FAQ - "even extremely uncommon disasters such as fires, tornados or flooding would only affect a single Avai
29. 539.▲
    by justinsb 14y ago

    I hope you blog more of these practices then. AWS doesn't put this stuff in writing, which is very convenient for them when something goes wrong, but makes it nigh on impossible to build a reliable system on EC2. I don't think it's an easy
30. 540.▲
    by justinsb 14y ago

    Source? AZs were supposed to be independent; they aren't. Fool me one...