wulczer

6 ms·

1.
by wulczer 11y ago
But that's because IBM has a strong presence in the SQL standard committee and can nudge it to standarise (post-factum) things they already implemented in DB2.
2.
by wulczer 11y ago
Since all of the previous states are there in the repo history, you can make that gif yourself at any moment!
3.
by wulczer 11y ago
Postgres at least will store rows roughly in insert order. So whether you're using integers or UUIDs for primary keys won't affect on-disk layout. When scanning the index, the database will be traversing it in btree order, anyway,
4.
by wulczer 12y ago
Sorry, ticked off my OCD :) It's better to say "key text" instead of "key varchar(1024)" and primary key already implies not null.
5.
by wulczer 12y ago
Thanks! We're staying in Spain and there are plans to further grow the Barcelona office.
6.
by wulczer 12y ago
Obligatory Show HN where we first showed Ducksboard to the world (1301 days ago): https://news.ycombinator.com/item?id=2332464 Thanks for all the feedback and interest we got from the community through all those years and
7.
by wulczer 12y ago
What is the epiphany here? What is the insight or enlightenment to be gained from this story? I have my own, but I'm curious about what people believe is the moral of the story.
8.
by wulczer 13y ago
That's also why, when you go to the movies and it turns out tickets cost the same regardless of the film, you should go for the one with the longest running time.
9.
by wulczer 13y ago
One of the important reasons Postgres connections are serviced by a separate process is that if that process segfaults, you only disconnect other connections (because if might have tainted shared memory when it died), but the main database
10.
by wulczer 13y ago
Almost two years ago the "The Pulse of Spain" was made, based on the same idea (disclaimer: the designers of this are also designers for our product (and dear friends)). It's in Spanish, unfortunately. You have to click on th
11.
by wulczer 13y ago
Since everyone is plugging their own: here's a identity provider daemon written in Go: https://github.com/wulczer/persona-idp It uses my Mail Transfer Agent to identify, so I can just use me email password to log
12.
by wulczer 13y ago
http://xkcd.com/447/
13.
by wulczer 13y ago
Logs subpoenaed from your ISP show that you were using Tor between 3:10 and 5:23 PM on 02/05/2012. Logs from forensic analysis of a breakin to EvilCorp show that the attacker came in from Tor and was downloading secret data from 3
14.
by wulczer 13y ago
Why can't you use a connection that has issued LISTEN for anything else? In fact, I'm fairly sure you can use it normally. We're using NOTIFY to broadcast database changes to services that should react to them and it's
15.
by wulczer 13y ago
Reminds me of http://www.supersimplestorageservice.com/
16.
by wulczer 13y ago
BTW, I think at least in Peninsular Spanish you'd say "Viva Brasil" (omitting the article).
17.
by wulczer 13y ago
We wrote a similar utility for our ORM-for-SaaS library some time ago, using lxml if available and falling back to ElementTree: https://github.com/ducksboard/libsaas/blob/master/libsaas/xm... Bear i
18.
by wulczer 13y ago
Not really, the ones calling down to date_trunc should be marked as STABLE, since the underlying function is STABLE. psql=$ \df+ date_trunc Schema | Name | Result data type | Argument data types |
19.
by wulczer 13y ago
I've been thinking about switching to sup a few times, but the inability to sync changes back to the IMAP server is a deal breaker to me. If I mark an email as read in sup, I don't want it to show up as unread on my phone. If I de
20.
by wulczer 13y ago
We have one on https://demo.ducksboard.com/ too :)
21.
by wulczer 13y ago
Doesn't matter if it's MD5 or SHA512. You can extend either hash function in the exact same way ( http://en.wikipedia.org/wiki/Length_extension_attack )
22.
by wulczer 13y ago
it does when using the re.U flag re.match(r'\d', u'੧', re.U) <_sre.SRE_Match at 0x3070ac0> sys.version 2.7.3 (default, Mar 4 2013, 14:57:34) \n[GCC 4.7.2]
23.
by wulczer 13y ago
What I meant is that using PKCS#5 padding ("N bytes of a value of N") will make it even more susceptible to a CBC oracle attack. I'm not nearly qualified to talk about that stuff, but instead of requiring PyCrypto maybe you should require K
24.
by wulczer 13y ago
Once this TODO is completed and real PKCS#5 is used, CBC padding oracle attack? https://github.com/mothran/bunny/blob/master/libbunny/AEScry...
25.
by wulczer 13y ago
I thought most disks nowadays support Secure Erase, so there's no need to futz around with dd. https://ata.wiki.kernel.org/index.php/ATA_Secure_Erase
26.
by wulczer 13y ago
Seems so, the code in question appears to have been added here: https://github.com/git-mirror/nginx/commit/7c7ad803a13d5f73b...
27.
by wulczer 13y ago
Create a new GitHub account, give it write access to the repo, generate a fresh RSA keypair, post the private key.
28.
by wulczer 13y ago
Check out http://www.gnu.org/software/stow/
29.
by wulczer 13y ago
Better idea: check which of those keys are in the Debian Weak SSH Keys list and own their GitHub accounts (or find if they have any servers and start SSHing into them). Less nefariously: send them an email asking to regenerate the key.
30.
by wulczer 13y ago
Yes, thank you, sorry for too big of a mental shortcut. By bad tests I mean trivial tests (much like trivial comments that just state the obvious), or ones that work on an assumption that need not be generally true and thus changes of the c

More ›