DrJosiah

26 ms·

181.
Show HN: Redis object mapper for Python
2 points by DrJosiah 14y ago|0 comments
182.
by DrJosiah 14y ago
Actually, no. What Diaz showed is that Beamr is offering nothing that actually improves either visual quality or bitrate; every setting that their method uses actually reduces quality compared to default x264 settings.
183.
by DrJosiah 14y ago
Everyone fat-fingers a database at some point... Then you build interfaces so that you can't make the same mistake.
184.
by DrJosiah 14y ago
No. And preliminary research suggests that they are easier to crack. http://arstechnica.com/security/2013/01/grammar-badness-make... Legitimately randomly generated passwords still win, as long as the server uses a salt with bcrypt or ite
185.
by DrJosiah 14y ago
You should watch the video linked in the parent post. A lawyer pays for themselves.
186.
by DrJosiah 14y ago
For what purpose are you writing software, and what is your desired destination? Do you want to program something interesting for you? (then figure out what you want to program, figure out the technologies it would take to make it happen, t
187.
by DrJosiah 14y ago
Codec registration has never been an issue. Let me repeat that with some emphasis, because it's an important point. Codec registration has NEVER BEEN AN ISSUE. And global registries are not inherently a bad thing. If you were to say "I don'
188.
by DrJosiah 14y ago
It's less convenient, it's a different API for every type of transformation, and the change has made code demonstrably worse. Further, the use of encode/decode is explicit. The only thing that was implicit was the automatic transformation o
189.
by DrJosiah 14y ago
The hex stuff is binascii.hexlify(). But I agree with you and with Armin; removing string -> string and unicode -> unicode encodings and decodings were a mistake. I said as much when the discussions about Python 3 and codecs were go
190.
by DrJosiah 14y ago
There are at least 2 better http servers for Python, uWSGI and gEvent. Tornado is known to be slower: http://nichol.as/benchmark-of-python-web-servers . More specifically, uWSGI has been shown to respond at under 25ms with 15k+ concurrent
191.
by DrJosiah 14y ago
I thought folks already knew about this? I know at least a few tech folks who have seen the Microsoft tablet demo'd, and been plied with free drinks to forget the experience. I can't believe that me, a tech guy in Los Angeles, heard about
192.
by DrJosiah 14y ago
I disagree with you on both points. You can't solve all (FS) problems for all people simultaneously. Custom FS solutions by Apple (recently), Oracle (20 years ago), and countless other vendors in the last 3 decades are about choosing differ
193.
by DrJosiah 14y ago
I know a couple people who moved from SRE to SWE at Google, though I didn't have a lot of exposure to Google proper (I was based out of Santa Monica), so I don't have numbers for testers to SWE. On the one hand, while I generally agree with
194.
Python bytecode hacks, gotos revisited(dr-josiah.blogspot.com)
5 points by DrJosiah 15y ago|0 comments
195.
by DrJosiah 15y ago
With caching, 512 intset entries, and 64 bit intset values, that's under 1 microsecond to intersect using the naive binary search algorithm for intersection. I don't believe that small sets are an issue for performance. For me, the real que
196.
by DrJosiah 15y ago
Let's say that you need to solve problem X. So you solve it with solution Z. People then come by later and say, "you should use A", or "B is better in that case", or even "C solves everything". The article was pointing out why A, B, or C wo
197.
by DrJosiah 15y ago
"Using C, I wrote a handful of lines of C code that took a sorted sequence of 4-byte integers that represented a Twitter user's followers, and I intersected that sequence against thousands of other similar sequences for other Twitter users.
198.
by DrJosiah 15y ago
With the work we were performing, the 2x improvement may not have been necessary, but it was useful. We could run everything on 1 box and keep up with roughly 30-50% system utilization continuously. Had we used the STL version, we would hav
199.
by DrJosiah 15y ago
But it's only for limited sized sets, 512 entries with the default configuration. Our sets were millions of entries.
200.
by DrJosiah 15y ago
I've updated the article to reflect the reasons why I pointed at vtables, which, as you say, are very likely incorrect. And no, the article does not say that Python was performing faster than the STL. In the first sentence in the "What I Bu
201.
Why we didn't use a bloom filter(dr-josiah.blogspot.com)
102 points by DrJosiah 15y ago|36 comments
202.
Creating a Lock with Redis(dr-josiah.blogspot.com)
8 points by DrJosiah 15y ago|0 comments
203.
by DrJosiah 15y ago
The title suggests some sort of study involving more than one person, the content isn't a study, it's an anecdote. Also, your "bottom line" is worse than patronizing, it's bullshit. I have friends that I talk movies with. I have other frien
204.
Improving Performance by 1000x(dr-josiah.blogspot.com)
1 points by DrJosiah 15y ago|1 comments
205.
Building for Uptime, Expecting Failure(youtube.com)
3 points by DrJosiah 15y ago|0 comments
206.
Building Adly's Twitter Analytics(dr-josiah.blogspot.com)
1 points by DrJosiah 15y ago|0 comments
207.
A Neat Python Hack? No, Broken Code.(dr-josiah.blogspot.com)
2 points by DrJosiah 15y ago|0 comments
208.
by DrJosiah 15y ago
Non-https images? It is not uncommon for browsers on https sites to not load http images.
209.
by DrJosiah 15y ago
A bit opaque, but cute. I've got a Redis + Python version (keeps the index in Redis) that I talked about last year: http://dr-josiah.blogspot.com/2010/07/building-search-engine...
210.
Building an Ad Network Ready for Failure(dr-josiah.blogspot.com)
2 points by DrJosiah 15y ago|0 comments

More ›