Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
zeendo
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
91.
▲
by
zeendo
7y ago
What would it even mean, though? It would just let them free up other, unencumbered funds, for things that aren't browser related when they would have otherwise been spent on the browser. Surely the influence of "wow look at all t
92.
▲
by
zeendo
7y ago
The parent is not only talking about the products they receive but how that money is spent. Those assurances are pretty impractical and I doubt any company can reasonably be held to that kind of standard.
93.
▲
by
zeendo
7y ago
Cellphones are cheap computers these days. It's very hard to participate in today's economy, much less its education system, without internet access.
94.
▲
by
zeendo
8y ago
Please don't make things up. It's detracts from the conversation.
95.
▲
by
zeendo
8y ago
I use the same setup as the GP. password-store uses git+gnupg so backing them up is a matter of distributing the git repo. My git repo lives on each of my laptops and I have a 'central' host of it...so it's backed up via dist
96.
▲
by
zeendo
8y ago
This is just a transparent attempt to rationalize / dig in to your original point right? :)
97.
▲
by
zeendo
8y ago
They're all over Atlanta. Midtown probably has the highest density of them but they're all over downtown, west midtown, Old 4th ward, Decatur, Reynoldstown, Cabbagetown, etc. Go down Memorial or DeKalb heading east in the morning
98.
▲
by
zeendo
8y ago
I'm also a cyclist but am pro-scooter...but, yes, I found that super disturbing. In fairness, the lifespan is 4 months. 3 months of profitability. Either way, it sounds crazy wasteful. Obvious questions arise (and answers are alluded t
99.
▲
by
zeendo
9y ago
The study doesn't differentiate between blowing through a stop sign and a rolling stop. You're drawing the conclusion you want, not what the data says.
100.
▲
by
zeendo
10y ago
It's (a little) more meaningful than your example-free claim. Your description of the startup scene here is literally the startup scene everywhere.
101.
▲
by
zeendo
10y ago
There are plenty of problems, but only two hard problems. Actually, I think there are three hard problems in CS. The third being pedants.
102.
▲
by
zeendo
10y ago
Would you suggest the Nix Pills series now? I notice it's 2 years old.
103.
▲
by
zeendo
11y ago
Again....source?
104.
▲
by
zeendo
11y ago
"go through whatever means necessary" sounds like questionable rhetoric to support your position. The difficulty involved in torrenting is quite low and there's arguably little moral reason to avoid doing so in this case sinc
105.
▲
by
zeendo
11y ago
...that's only true while the attacker controls the site. The 90 day window mitigates the effect of the cert being leaked. Just because I grabbed a site's cert on day doesn't mean I control the site. In fact, there's not
106.
▲
by
zeendo
11y ago
$VERBOSE is mentioned for warnings. That's pretty handy.
107.
▲
by
zeendo
11y ago
Umm. Shit. If you come up with a resolution to the cognitive dissonance you just exposed then please let me know.
108.
▲
by
zeendo
11y ago
Both of those solutions require Amazon to do "something" to the instance itself and thus limit the kinds of machines you can run on EC2. Custom kernels, FreeBSD, etc.. Their current metadata approach works across OSes. And, yes, t
109.
▲
by
zeendo
11y ago
Sure. My point is that avoiding merge conflicts has nothing to do with your branching strategy...it has to do with what work you parallelize.
110.
▲
by
zeendo
11y ago
Of course we still have merge conflicts. You do too. Or is your work just organized such that no two people are ever working on the same file concurrently? If it is then you'd not have any merge conflicts with trunk based development e
111.
▲
by
zeendo
11y ago
Or, say, Google and Facebook? http://paulhammant.com/2014/01/08/googles-vs-facebooks-trunk...
112.
▲
by
zeendo
11y ago
If everyone is working in feature branches then you're still merging less often. Until you've merged your work into master then I'm not merging it into my feature branch. That leads to bigger merges than if everyone is commit
113.
▲
by
zeendo
11y ago
In my experience, if you have high bandwidth communication (e.g. a team room) then trunk based development scales to well beyond 2-3 people. I've seen it work on at least one large distributed teams (3 team rooms, 3 time zones, ~30 dev
114.
▲
by
zeendo
11y ago
Keep in mind though that if everyone is working on something that takes multiple days to complete (and your definition of non-working is incomplete) then all the work is happening in feature branches and master is quiet. Those rebases are e
115.
▲
by
zeendo
11y ago
Probably iOS development?
116.
▲
by
zeendo
12y ago
What do you mean by 'isomorphic app' ?
117.
▲
by
zeendo
12y ago
Thanks for the heads up. I was about to sign up for the same reason. Have you found a better alternative for duplicity backups?
118.
▲
by
zeendo
12y ago
Package repositories go down too (Maven, RubyGems, NPM, etc...). Unless by 'lax' you mean 'didn't check in your dependencies'...I don't think I'd call that lax, though.
119.
▲
by
zeendo
12y ago
You say politically correct and I say treating people with respect.
120.
▲
by
zeendo
12y ago
I'm sorry that you think it's impossible to fix. I think it is possible to fix by getting some balance in our numbers. Getting more women around, basically. If we see more women in software then we'll start to get rid of our
More ›