Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Left-pad as a service

left-pad.io

Post reply on HN
Print view View original

231–240 of 269 posts

Re: Left-pad as a service

#231

Post by toyg »

mikeash wrote: post #205
toyg wrote: post #200

Earlier quoted context omitted.

The reason is maintenance. Once you set up yet-another-repo-mirror (after your apt, rpm, nuget, pypi etc mirrors), someone needs to keep it up, back it up, secure it, refresh the packages, etc etc. It's "cloud culture": rely on some else's maintenance effort and just work on your own problems. Like all things, it has drawbacks.

I don't buy it. You need a place to keep your own repository, and once you have that, keeping copies of other repositories is pretty much free. Keeping up, backing up, and securing N repositories is no more work than doing that for one repository. You can still outsource this, even. There's nothing wrong with using GitHub or Bitbucket or whatever for your repositories, just make sure that you're using your repositori…

You need a place and a person who knows all this stuff (or likely more than one). That's expensive and hard, as knowledge of all these setvices has to be correctly handed over across people and over time. That's on top of org-specific processes that are likely more important and more byzantine, which means they have priority. So the day your awesome_pkg_server breaks, nobody remembers how and why it was set up, and the whole thing is scrapped. Back to square 1. This assuming you actually have one or two people with enough free time in the first place...

Whereas standard "fetch package" scripts are trivial, and each individual stack is known well enough by developers working on it every day that there is zero administration and little need for extensive knowledge transfer... At least until left-pad disappears from under your feet; but that's rare enough that the trade-off is worth it overall (or people cannot even imagine it ever happening).

I'm not saying this is how it should be, I'm just describing why a lot of people do what they do.

Maybe the solution is an overall simplification and harmonization of all these services, so that the responsible option and the lazy option won't differ so much. You could have something like a prebuilt "repo server" image that can be downloaded and configured quickly, and which will then auto-cache all requested packages across different services. Cache invalidation would still be tricky but could be triggered selectively from an admin panel. Maybe all this should be packed into CI servers by default...

Re: Left-pad as a service

#233

Post by pfooti »

TeMPOraL wrote: post #214
pfooti wrote: post #126

Earlier quoted context omitted.

I'm pretty sure that attitude is called "paternalism". Self-regulation works better when it's self-directed, as opposed to other-directed (as in, someone else doing the regulation). `I shouldn't do ${foo} because people will make fun of me` is a horrible attitude to foist on anyone else, and anyone who tries to propagate that attitude isn't helping the community at all.

> `I shouldn't do ${foo} because people will make fun of me...` `... but why are they laughing from ${foo} in the first place? They're smart people, so there might be a reason. Let me research it more first...`. This is how self-regulation works.

I think there's a fundamental difference between, say [0] (thoughtful, nuanced), and [1] (just plain silly). The initial posts are all worthwhile and interesting, with the latter clearly being satire. It would be even funnier if the OP exposed a more robust string-handling library from ruby or python to the javascript runtime in node by way of a native lib binding. [2] is kind of just elitist, "get off my lawn" yelling, which misses the point fundamentally in this problem.

There are different flavors of comments on these threads. Some are pretty cogent and thoughtful. Others aren't. A "sky is falling augh, isArray is one line, javascript devs are teh worst" misses the underlying problem of javascript being addressed (interpreter fragmentation, not having a reliable way to test if something is an array on older platforms). I'd argue it doesn't contribute at all to the improvement (metacognitive or otherwise) of any javascript dev, but instead serves to harden people's opinions on one side or another of a mythical line.

The difference is whether the goal is to contribute to the discourse or just go for the shallow laughs. It's an important difference, because the former does lead to improvement all around, whereas the latter leads to high-fives and community decay.

0: https://news.ycombinator.com/item?id=11352704 1: https://news.ycombinator.com/item?id=11350651 2: https://news.ycombinator.com/item?id=11348798

Re: Left-pad as a service

#234

Post by pmlnr »

faizshah wrote: post #24

I don't understand why this community has to have a weekly cycle of bashing different programming communities. Every week there's a new drama thread bashing Java devs, Go devs, Javascript devs etc. The thing that I come to this community for every week is to read about new developments in our industry, if you don't come here for that then what are you coming here for? And wasn't it just a few months ago people were p…

This is not new, and not limited to HN. in my opinion we need to be reminded that all tools have flaws.

https://wiki.theory.org/YourLanguageSucks

Re: Left-pad as a service

#235

Post by stcredzero »

TeMPOraL wrote: post #217
smarks wrote: post #206

Earlier quoted context omitted.

Exactly. That's the No True No True Scotsman Fallacy Fallacy.

Compare also with Fallacy Fallacy - just because the argument is fallacious, doesn't mean the overall point is wrong.

There's this "Tone Argument" Fallacy Fallacy. You can't win an argument by complaining that the other's tone is unpleasant. So that's a valid fallacy, but it only applies to a particular argument in a particular circumstance. Certain people on the Internet then make the illogical jump to the conclusion that they are therefore justified to use whatever tone they care to in any argument. (And lo and behold, they always "win" arguments.)

Re: Left-pad as a service

#236

Post by uremog »

8note wrote: post #228
talles wrote: post #188

Earlier quoted context omitted.

Jesus Christ, is that a real thing!? http://docs.spring.io/autorepo/docs/spring/2.5.x/api/org/spr...

> Convenient proxy factory bean superclass for proxy factory beans that create only singletons. that actually makes sense to somebody?

At least it's convenient...

Re: Left-pad as a service

#237

Post by haldean »

8note wrote: post #228
talles wrote: post #188

Earlier quoted context omitted.

Jesus Christ, is that a real thing!? http://docs.spring.io/autorepo/docs/spring/2.5.x/api/org/spr...

> Convenient proxy factory bean superclass for proxy factory beans that create only singletons. that actually makes sense to somebody?

Not sure what about that is hard to understand, it's a convenient proxy factory bean superclass for proxy factory beans that create only singletons. Says it right there on the page.

Re: Left-pad as a service

#238

Post by kdma »

jameshart wrote: post #36
c4n4rd wrote: post #14

This is really exciting!!! I was a bit disappointed that the right-pad will be out only in 2017. I am looking forward to that release because there is a high demand for it now. What kind of load balancing is being used on the back-end? I called leftpad(str, ch, len) with the length I needed and noticed that is not very scalable because it is blocking. A better approach I would recommend to those using it is to call t…

Best practice for performance for large left-pad jobs is to call the service recursively using mapreduce. Remember that left-pad(str,pad,n) is equal to left-pad(left-pad(str,pad,n/2),left-pad("",pad,n/2),n). This should run in logarithmic time and is highly parallelizable. If you don't like the "" magic string in there you could replace it with a call to left-pad(null, null,0).

Shouldn't it be leftpad(leftpad (str,pad,n/2),leftpad("",pad,n/2),1)?

Re: Left-pad as a service

#239

Post by acqq »

TeMPOraL wrote: post #224
acqq wrote: post #57

Earlier quoted context omitted.

I don't consider C++'s Boost something to be proud about. Last time I've checked, Google forbade a significant part of it to their developers, and in my opinion rightly so. But who likes everything there should just as well understand this Node affair.

Google makes a lot of controversial decisions from engineering standpoint, not least of which is doing half of their stuff in Java. (It always bewildered me - why hire the best of the best, and then make them work with lowest-common-denominator tools?)

From my point of view, if we remain by why not allowing everything Boost, it isn't controversial at all, for the same reason that Linus (an I too) believe that it is a good that Linux kernel is pure C and not C++.

And the same reason why some people now see that the presented Node popular practices aren't something that the battle-tested players consider reasonable to do.

Re: Left-pad as a service

#240

Post by B1FF_PSUVM »

mschulze wrote: post #173

As a Java developer, I am a bit jealous. When people joke about us we usually only get a link to the Spring documentation of AbstractSingletonProxyFactoryBean (or maybe the enterprise hello world), but no one ever wrote that as a service. Maybe someone can do that? https://abstractsingletonproxyfactorybean.io seems to be available!

> When people joke about us we usually only get a link

I take it you're not familiar with the Bile Blog, e.g.

http://www.jroller.com/fate/entry/rip_jboss_mail_services

Alas, service ceased ten years ago.

Post reply on HN