Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
dickeytk
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
35 ms
·
331.
▲
by
dickeytk
14y ago
http://www.moo.com/blog/2012/09/27/the-business-cards-of-the...
332.
▲
by
dickeytk
14y ago
I think being 'easier' in a Rails app is debatable. Facebook authentication certainly isn't 'hard'. IMO OpenID is an awful user experience, I much prefer using Facebook. But that point is certainly debatable as well.
333.
▲
by
dickeytk
14y ago
I wrote the aforementioned article. Being an intern implies a temporary/part-time position or a student position, however the dictionary term is: "A student or a recent graduate undergoing supervised practical training." I think that's abso
334.
▲
by
dickeytk
14y ago
The difference between that and bootcamp is that the end-goal used to be to get a certificate that in theory could get a job. Bootcamp is different because the goal is for the engineers to be bootstrapped to the point they can use google to
335.
▲
Dev Bootcamp is the future of our industry
(dickey.xxx)
51 points
by
dickeytk
14y ago
|
19 comments
336.
▲
by
dickeytk
14y ago
It also seems to be the flakiest feature
337.
▲
by
dickeytk
14y ago
I really wish I could agree with you, but in my experience, ActiveRecord is just way easier to deal with than Mongoid.
338.
▲
by
dickeytk
14y ago
except it doesn't, it's an optional extension
339.
▲
by
dickeytk
14y ago
RDS is awful. The interface is shitty and buggy, offers graphs that you can't do anything with, the backups cause our system to fail every night, EBS backing has serious spiking problems, you can't access the replicas and it runs a shitty c
340.
▲
by
dickeytk
14y ago
where does postgres need to catch up? and what knowledge gap?
341.
▲
by
dickeytk
14y ago
Postgres has had solid replication since 8.4
342.
▲
by
dickeytk
14y ago
I should've noted in the article, almost all of those are optional extensions to Postgres
343.
▲
by
dickeytk
15y ago
So I know for a fact that disqus can export into XML, not sure about livefyre, but... you probably know how I feel about them after the article haha
344.
▲
Commenting systems: Disqus vs Livefyre
(jeffdickey.info)
2 points
by
dickeytk
15y ago
|
2 comments
345.
▲
CloudFlare review: not snake oil, but it feels like it
(jeffdickey.info)
1 points
by
dickeytk
15y ago
|
0 comments
346.
▲
by
dickeytk
15y ago
I don't think you were at all. Many people mentioned the same thing you did. I'm new at blogging, so I'm trying to take in everything and produce the best content I can, so I love the feedback! I would love to see a follow-up article too by
347.
▲
by
dickeytk
15y ago
Your understanding is correct, but those 'other features' is right where Redis becomes interesting. If you just want a better memcache, use membase.
348.
▲
by
dickeytk
15y ago
tabby, I've also updated the article to mention the fact that KEYS may not be a solid use of the system. I would appreciate any feedback!
349.
▲
by
dickeytk
15y ago
doh! I got it backwards. You are right that hashes are more efficient: http://redis.io/topics/memory-optimization However, for this particular problem, the docs do say: Don't use KEYS in your regular application code. If you're looking fo
350.
▲
by
dickeytk
15y ago
YES! It's a winning powerhouse.
351.
▲
by
dickeytk
15y ago
Yes!! Thank you so much. That's exactly the point I was trying to make.
352.
▲
by
dickeytk
15y ago
It's no authoritative source, but I'm obviously not the only one that thinks this: http://www.webdevelopment-blog.com/2011/07/puma-on-redis/
353.
▲
by
dickeytk
15y ago
Thanks for linking the antirez article, I actually haven't seen that one before I wrote the article. I put a link to it in the article. As far as they keys thing, it's obviously been a major point of contention with my article. I know the w
354.
▲
by
dickeytk
15y ago
Right, I know what the docs say, however, it's fine to use this for cache invalidation (but maybe not other use-cases) in production. The docs also say: "While the time complexity for this operation is O(N), the constant times are fairly lo
355.
▲
The Great Redis Misapprehension
(jeffdickey.info)
65 points
by
dickeytk
15y ago
|
40 comments
356.
▲
by
dickeytk
15y ago
Codes are just randomly generated then stored in a postresql db for later reference. If there is a collision, it adds a digit.
357.
▲
by
dickeytk
15y ago
I guess Twilio has UK numbers in beta right now. Maybe I could clone the site and throw it up on another Heroku instance and use one of those.
358.
▲
by
dickeytk
15y ago
It appears to me (in my 1 minute look at the site) that it's a server-side framework. backbone and knockout are client-side EDIT: I think what I just said is incorrect since the article makes a backbone comparison
359.
▲
by
dickeytk
15y ago
Trouble is, without accounts, how can I tell who the originator of a conference is? My goal is to keep the UX as brutally simple as possible. I'm really against having payments get in the way of using the product in ANY way since I feel tha
360.
▲
by
dickeytk
15y ago
No problem! I MIT licensed it as well, so feel free to hack, change, sell away! If you have any questions about it, let me know.
More ›