kmavm

23 ms·

181.
Finally, a Real Creator is Person of the Year(mises.org)
20 points by kmavm 16y ago|21 comments
182.
Tunisia invades, censors Facebook, other accounts(cpj.org)
17 points by kmavm 16y ago|6 comments
183.
Facebook's Exceptional Approach to Vulnerability Disclosure(eff.org)
1 points by kmavm 16y ago|0 comments
184.
by kmavm 16y ago
Not sure why this suddenly hit the front page of HackerNews, but, uhh. I wrote this. AMA.
185.
What's wrong with OpenID?(quora.com)
2 points by kmavm 16y ago|0 comments
186.
by kmavm 16y ago
I interpret the graph as saying that Google's share is in fact the smallest; this makes sense when you consider that it is display advertising. The missing 50% is not more Google, but long-tail.
187.
by kmavm 16y ago
Here's what Google had to say last year about exporting emails from their social network, Orkut: "Mass exportation of email is not standard on most social networks — when a user friends someone they don’t then expect that person to be easil
188.
Google on why Orkut should not export email contacts (10/26/09)(techcrunch.com)
4 points by kmavm 16y ago|1 comments
189.
by kmavm 16y ago
Here's what Google had to say about social networks and email exporting less than one year ago: "Mass exportation of email is not standard on most social networks — when a user friends someone they don’t then expect that person to be easi
190.
by kmavm 16y ago
> This will slow down my AOL-using friends who gave away all their contact info to Facebook and now I get pelted with spam ... Right; there are practical usability problems with allowing willy-nilly access to your contact list. Some of
191.
by kmavm 16y ago
Sorry for any perceived "valley-centrism"; feel free to substitute "small industry" for "small valley." What does Zuck's level of personal awesome have to do with anything? This strikes me as a major confusion about the business/consumer re
192.
by kmavm 16y ago
> less intentional. I'd refer you to Hanlon's Razor. You might also want to introspect as to why the singular social privacy fiasco whose decision-making you've seen internally strikes you as less evil than the privacy fiascos you've wi
193.
by kmavm 16y ago
Not to speak for GP, but that is a perfectly reasonable position to take, and one that I would take. They're different services; GMail does not have third-party apps in the same sense. Facebook's app ecosystem has some quite disreputable pa
194.
by kmavm 16y ago
Again, Facebook exports a unique identifier for each friend already, and has been doing so for years. This is the entire point of Facebook Connect.
195.
by kmavm 16y ago
Laugh while you can. During the Buzz launch, I recall Google found it just as hard to negotiate the intrinsically tricky balance between privacy and openness. There is a real trade-off between the two in products that allow you to share inf
196.
by kmavm 16y ago
The graph API provides exactly what you are asking for. It doesn't provide friends' names , but FBIDs. http://www.facebook.com/profile.php?id=<id&#62 ; is the user's profile, and http://graph.facebook.com/<id&#62 ; is the public
197.
by kmavm 16y ago
Would anybody like to actually read the words, understand them, and engage with the argument they contain, rather than snark about how FB == teh hatez? The meat of it is: "Each person owns her friends list, but not her friends’ information.
198.
by kmavm 16y ago
Google is specifically demanding an API , and further making demands about performance, uptime, availability, etc. of the API. They're being perfectly clear that exporting the emails to .csv file that then gets uploaded is unacceptable. An
199.
by kmavm 16y ago
They are different services . Facebook is a third-party platform in a way that GMail has never been. If your email address becomes your friends' Facebook data, your email address will be distributed to many companies that you have no d
200.
by kmavm 16y ago
Google is, imho, perfectly within its rights to provide an email service that doesn't allow you to download your contacts easily, or doesn't allow it for certain services, or whatever. What I don't appreciate is the hypocritical Monday-morn
201.
by kmavm 16y ago
Some points that are getting lost in the discussion below. I work at Facebook. 1. Facebook lets you export your data. It has been possible to do so ever since the graph API debuted in April '10. Since the market wasn't filling the gap, we e
202.
by kmavm 16y ago
It is perfectly easy for other web services to access the data of a logged-in Facebook user, provided that the user has granted permission via Facebook Connect. This is how Facebook's "Download your Data" product[1] works; it is entirely im
203.
by kmavm 16y ago
In the parlance of Facebook critics, robots.txt is "opt out." But, if you don't like my web data example, consider streetview data. I don't recall asking Google to build a fleet of androids to drive around my neighborhood taking pictures
204.
by kmavm 16y ago
'if Google accessed some other sites data but didn't provide a reciprocal arrangement, then I would be equally harsh on Google.' This is exactly what Google does: read the open data that all other sites provide, over http, add a lot of valu
205.
by kmavm 16y ago
By "any metric"? How about, "lines of open source contributed per engineer-year"? Google is 10 times larger than Facebook, and has been around for twice as long. You would expect it to have a longer list of open source contributions no matt
206.
by kmavm 16y ago
Alas, my understanding of this point is entirely folkloric. Self-experimentation, and working on improving real interactive systems, agrees with it. Please update if you find anything more comprehensive.
207.
by kmavm 16y ago
A query that is always slow will not end up on home.php. A query that is slow 1% of the time might.
208.
by kmavm 16y ago
For what it's worth, Jeff Dean of Google has famously emphasized 95th and 99th percentile performance in preference to average or median performance over the years. The realization that edge cases more powerfully determine user perception o
209.
by kmavm 16y ago
True enough in 1999, and all but completely false now. For a competently implemented hash map with integer keys, hash tables start beating linear search somwhere between n = 2 and n = 4. At n = 4, you start needing more than one cacheline t
210.
by kmavm 16y ago
Even before Chrome, they know what you read on a very wide cross-section of the web, too, thanks to the AdSense pixel.

More ›