gleb

21 ms·

211.
by gleb 15y ago
It's not about being disjoint, you need to known total numbers to make a useful conclusion. It's not really that the title is wrong, it's just misleading. In the same way as the following is misleading: Gentiles far more likely than Jews
212.
by gleb 15y ago
Their data doesn't support that conclusion. Actually, you can't make any conclusion without knowing total number of MSinE vs MBAs. Update: http://www.cgsnet.org/portals/0/pdf/R_ED2008.pdf (p.16) suggests there are a lot more business ma
213.
by gleb 15y ago
Very nice. Used it to research upcoming purchase quickly, would have signed up to persist saved results if you had that option. Main feedback - browser refresh, back/forward, tabs, middle-click should work, e.g. - refresh now loses half th
214.
by gleb 15y ago
Patrick, rdouble is not being nice, but his major point is right. There are different forces in play for negotiating engineer's salary than a consulting engagement. Employee compensation within a company needs to be equitable. Similar work
215.
by gleb 15y ago
Real estate purchases in US are usually highly leveraged, say 5x for a typical purchase with 20% down. So, negotiating those will have proportionally larger impact than immediate cashflow changes suggest. The article incorrectly suggests ot
216.
by gleb 15y ago
There is similar problem in Postgres as well. Not nearly as bad in magnitude, and with some shared responsibility with Linux and file system code (since PG relies of filesystem's caching). Greg Smith talked about it in some detail during
217.
by gleb 15y ago
Not every developer has a brand. This service appears to target companies hiring developers that don't. Fortunately that is 99.99% of the market.
218.
by gleb 15y ago
Having write disk cache on would certainly explain it. But that leaves the question of discrepancy with numbers with competitors. You turn off write-through caching on disks when you run a database unless you are willing to accept corrupti
219.
by gleb 15y ago
The synchronous writes benchmark is interesting. This is normally bound by # seeks your disk can do per second, which is mostly a function of rotational speed. With 7200RPM drive you get 7200/60 = 120 of these a second. So the 100 and 110
220.
by gleb 15y ago
Congrats on the launch!
221.
by gleb 15y ago
That's hilarious. Unfortunately trademarks are phonetic, so this can't fly.
222.
Apple legal is going after Open Source startup with their App Store trademark(crunchgear.com)
20 points by gleb 15y ago|12 comments
223.
by gleb 15y ago
Congrats on the launch!
224.
by gleb 15y ago
PageRank for karma, interesting idea.
225.
by gleb 15y ago
I think improvement in civility that no-comments brought is a local maximum. I hope so anyway. It's worth a try to experiment with other options. Like charging karma for commenting. And maybe try these as A/B tests. Half the topics run wi
226.
by gleb 15y ago
I no longer read comments. Points acted as an editor's cue on high quality content to read. Maybe more good comments are produced now, but it's no longer possible to find them without reading everything.
227.
by gleb 15y ago
Today, they would have applied with a better idea, in a later stage of implementation.
228.
by gleb 16y ago
If comments cost karma it would lead to less discussion, which I think would lead to better discussion. Popular comments will make more karma than they cost, so users will still be encouraged to leave comments that will become popular. It s
229.
by gleb 16y ago
I'd try to severely decrease total # of comments. Really bad comments are not the root of the problem. Simply having large number of mediocre comments crowds out and discourages thoughtful discussion from starting at all. I'd say: * create
230.
by gleb 16y ago
He can do what he wants to do for free. See "same day standby" and "confirmed standby": http://www.flyertalk.com/forum/united-mileage-plus/955496-un...
231.
by gleb 16y ago
This kind of survivor bias will be true for mostly any subscription business. Users that weren't getting much value from your business will drop off. Given a reasonable segmentation of free and paid users that will mean a larger percentag
232.
PostgreSQL Conference - November in San Francisco(postgresqlconference.org)
2 points by gleb 16y ago|0 comments
233.
by gleb 16y ago
This is awesome work. I've been following these commits on GitHub and it's great to see AR being optimized. Unfortunately there is not enough info in these benchmarks to tell whether for a large production app 3.0 might be 10 times slower
234.
by gleb 16y ago
Acunote ( http://www.acunote.com ) - project management, wiki, issue tracking, scrum software Tarsnap ( http://www.tarsnap.com ) - offsite backup DnsMadeEasy ( http://www.dnsmadeeasy.com ) - DNS TrustCommerce ( http://www.trustcommer
235.
by gleb 16y ago
I'd pay for Kindle synchronization to actually work. Mine hasn't in a week or so.
236.
by gleb 16y ago
It would be helpful to elaborate on: * what document.domain is * what hijacking user's session actually implies
237.
by gleb 16y ago
Ugly it definitely is.
238.
by gleb 16y ago
K9 is good. Not up to the standards of 6 year old ChatterEmail on Palm just yet, but vastly better than the builtin client of iPhone's Mail. And it's an active Open Source project improving on original Android's source, a direct benefit of
239.
by gleb 16y ago
Joost failure had nothing to do with technology. Instead they fundamentally misunderstood movie business and had a wrong business model. They did solve half of the problem, cost of bandwidth, but they failed to solve the much harder half -
240.
by gleb 16y ago
We use and maintain this: http://www.acunote.com/open-source/javascript-keyboard-short... It's modeled on Google Experimental, and other Google/vi-style shortcuts. Would love to see something like this implemented in HN codebase.

More ›