Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
keefe
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
64 ms
·
661.
▲
by
keefe
17y ago
I'm thinking... reasonably difficult to crack
662.
▲
Ask HN: Insights into securely distributing binaries (call home)?
2 points
by
keefe
17y ago
|
4 comments
663.
▲
by
keefe
17y ago
It depends on how close you are to completion. How much of your codebase can be repurposed to a less populated niche?
664.
▲
by
keefe
17y ago
I should probably go to the password generators, I have a series of passwords I use now with any important (banking) site having a unique password.
665.
▲
by
keefe
17y ago
it's just the definition of the word, right? You presumably hire someone with a set of responsibilities - some mix of write this software, design this UI, optimize this server and you're at some degree of freedom on the requirements. Set hi
666.
▲
by
keefe
17y ago
I'm a big fan of document databases. It was one of those things where I was working on my app, thinking to myself... self, aha! I need documents or arbitrary kvp storage... then thinking, yeah somebody else must have done that already and t
667.
▲
by
keefe
17y ago
the thing that pisses me off most is... I sign up for some website and a few minutes later I get some email - hey welcome to site X your username is Y and your password is Z... WHAT THE FUCK
668.
▲
by
keefe
17y ago
I think you must assume it's not secure on most sites and use a "public" and "important" password.
669.
▲
by
keefe
17y ago
damn right. Totally stupid idea - perhaps they should revise their expectations to something more realistic?
670.
▲
by
keefe
17y ago
my neural networks, I train them....... depending on how much time I had with it, I'd work on additional networks maybe something on stock data
671.
▲
by
keefe
17y ago
it's also a WoW instance
672.
▲
by
keefe
17y ago
thx I'll have to check those out, my mathematica license is gone :[
673.
▲
by
keefe
17y ago
any chance at all these direct comments were one of those jokes I've been hearing so much about lately? It's hard to believe that a successful person would have such a bias for round numbers rather than state changes - certainly there is a
674.
▲
by
keefe
17y ago
lol I'm 29, I better apply!!! I'm gonna have to remember to point out that I registered my domain and started work on my idea when I was only 25
675.
▲
by
keefe
17y ago
I'll just throw this out there because it's not in the mainstream, but Mathematica is well worth considering if you are still @ uni and can get it for cheap. The advantage is that mathematica is a very powerful language for doing "mathy" st
676.
▲
by
keefe
17y ago
Title is a bit misleading - sure 20 min a day may be fine for achieving some milestone, but the bottom line is that our bodies are machines that are not meant to be sedentary!!! at the very least, 1 lb of fat = 3200 calories and we should n
677.
▲
by
keefe
17y ago
parallelization is not only about scalability, but also performance. If a task takes 20min and can be parallelized 4X, it now takes 5 minutes. There's only a certain portion of each program that is inherently sequential. We're at 2-4 cores
678.
▲
by
keefe
17y ago
Exactly what type of application are you referring to? Most server software does a lot of work in parallel. Many computationally complex algorithms are parallelizable to one degree or another. Video games usually have a lot of parallelizabl
679.
▲
by
keefe
17y ago
you've actually got the processing power to do interesting things. I've currently got a 250M record database in my domain of interest - a few years ago, crunching on this database was prohibitively time expensive but now it flies, without e
680.
▲
by
keefe
17y ago
first, you need to grok the basics http://academicearth.org/courses/machine-learning but data mining can't be learned from a passive standpoint. You need to find a large dataset http://aws.amazon.com/publicdatasets/ and try to do someth
681.
▲
by
keefe
17y ago
I worked at one of the bigger semweb startups for a few years, been to the major conferences several times... the whole concept of putting structure to data makes a lot of sense. However, an explicit ontology is not necessary to make data u
682.
▲
by
keefe
17y ago
it's not going to pan out because we're going to roll back to single cores?
683.
▲
by
keefe
17y ago
I'm not sure they so much loaded the game as deliberately circumvented the Warden http://en.wikipedia.org/wiki/Warden_(software) to copy portions of the WoW memory space (locations of toons, etc) into their own address space... but I thin
684.
▲
by
keefe
17y ago
I used to run daily and I have chronic problems with my knees stemming from a childhood accident. Anecdotally, I have much less pain using an elliptical machine.
685.
▲
by
keefe
17y ago
I have similar habits, which feel like a big time leak. I recommend git because I find the peer to peer model more elegant, but I've yet to use git collaboratively.
686.
▲
by
keefe
17y ago
insufficiently currently : master git repository for everything, something like ~/docs lots of breakdown of documentation more git repositories under : ~/docs/code/projects - stuff I am currently working on or using ~/docs/code/legacy_pro
687.
▲
by
keefe
17y ago
My dad left his old Atari (2600 Junior, I think) in my room and I found basic on there towards the end of grade school. This was really frustrating to use, so into middle school I would coerce my dad into getting up an hour early so I coul
688.
▲
by
keefe
17y ago
"If our newly-independent programmer does not make 10x the money, we must admit that she is not the 10x programmer we thought she was." Not a good metric. The signal is at least polluted by social skills and geography - in some areas of the
689.
▲
by
keefe
17y ago
launch my 3 year old web app to the public and watch the data flow from a pile of people using it. Get the code efficient enough that the pile doesn't break the bank in server costs.
690.
▲
by
keefe
17y ago
eclipse default because I don't particularly care
More ›