Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
je42
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
34 ms
·
331.
▲
by
je42
9y ago
I wonder why they want to do the step: - Followed by removing Nginx + Memcached couple altogether out of equation.
332.
▲
by
je42
9y ago
This PDF is pretty nice. Regarding the differences of Kanban vs Scrum. https://www.crisp.se/file-uploads/Kanban-vs-Scrum.pdf
333.
▲
by
je42
9y ago
in a state less auth layer you can't do blacklisting of token. hence if your security requirement are somewhat in the "safer" ranger. you will need to use a bit of state to whitelist or blacklist tokens.
334.
▲
Cross-platform, graphics API agnostic, BYO-engine style rendering library
(github.com)
2 points
by
je42
9y ago
|
1 comments
335.
▲
by
je42
9y ago
and boy if it doesn't mirror org structure. you get pain :)
336.
▲
by
je42
9y ago
also this: https://www.slideshare.net/RandyShoup/goto-aarhus2014-enterp...
337.
▲
by
je42
9y ago
multiple services per team are trickier. also depends on their complexity. sometimes you can't get around it. it also depends on your infrastructure to manage services. if setting up, managing, deploying and monitoring a service is che
338.
▲
by
je42
9y ago
yeah good to point to make it explicit. splitting to smaller (but not micro) is a very good transition, that might already help.
339.
▲
by
je42
9y ago
1) Agreed, the split into libraries can be fine. At some point, you want teams to deploy independently of other teams. 2) agreed ! thanks for adding the qualifiers !
340.
▲
by
je42
9y ago
it depends on your reasons. - if you need to scale the number of teams working on the same product you need to introduce microservices to make the teams work independently from each other. - if you need to scale the (monolith) service and i
341.
▲
by
je42
9y ago
90% developer tests here
342.
▲
by
je42
9y ago
we just went for aiohttp. django channels didn't scale on even very simple benchmarks. while aiohttp didn't even sweat. that said, making everything async in the python sync world was some effort. ;)
343.
▲
by
je42
9y ago
almost no test coverage. did i miss them ? for proper use in production you would need to have hundreds of unittests and a whole bunch of component + integration and e2e tests.
344.
▲
Ask: how do you test react-components against multiple versions of react?
1 points
by
je42
9y ago
|
0 comments
345.
▲
by
je42
9y ago
very good book. but it is also very old. why are they using it only now ???
346.
▲
by
je42
9y ago
don't forget that full internationalization also includes taking care of layouting issues i.e. words that are longer in other languages. Characters that are higher. Right to left text.
347.
▲
by
je42
9y ago
Lot's of people mentioning "reactivity as being expensive/non-scalable". is there a good read up on what reactivity in the context of meteor.js is and why it is expensive ?
348.
▲
by
je42
9y ago
the article mentions, that HAProxy cannot do a reload without downtime ? is that correct ? no way around that ?
349.
▲
by
je42
9y ago
it's an implementation detail. not guaranteed. :( which makes this even worse.
350.
▲
by
je42
9y ago
how does this compare to Space-X Raptor engines ?
351.
▲
by
je42
9y ago
you can run a redis cluster.
352.
▲
by
je42
9y ago
actually, you should see the pain that you get when having unordered dictionaries by default. which is the default in python, you get awesome effects, where the order of the unordered dictionary is dependent on the random seed. where librar
353.
▲
by
je42
9y ago
actually, i am disappointed. if someone posts a library and then doesn't have any tests. It will be causing pain for the guys/girl how are using it. i.e. minor changes. that are actually breaking etc... On top, no tests => inf
354.
▲
by
je42
9y ago
why down vote ? do you really want to use a library without tests ?
355.
▲
by
je42
9y ago
no tests. :(
356.
▲
by
je42
9y ago
why down vote ? fork is fine as long as the communication between the two processes is low-bandwidth.
357.
▲
by
je42
9y ago
fork works fine in node.js
358.
▲
Using Istio in production?
1 points
by
je42
9y ago
|
1 comments
359.
▲
by
je42
9y ago
Interesting . We are calling this approach internally: component-based. Other names, i heard were widget based.
360.
▲
by
je42
9y ago
OMG. MFC ? You mean Microsoft Foundation Classes ? mmh. that stuff is still supported ? eventhough there was/is all the .NET crazyness. That's actually pretty good of MS. I guess the main question is: how do you in advance if you
More ›