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

Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

answers.microsoft.com

Post reply on HN
Print view View original

281–290 of 492 posts

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#281

Post by bubblethink »

skybrian wrote: post #215
pbhjpbhj wrote: post #210

Earlier quoted context omitted.

So they make it run buggy in order to reduce the support burden??? Surely that makes for more support calls. I'd have much more respect for them if they just came clean and said "we hate linux users, if you want to use that OS don't use our products because we try our best to make them not work on linux". Look at what they did with Skype for example. At least then we'd know not to try their stuff.

Almost always, nobody has any intention to "make it run buggy." It's about making it run well on platforms you care about (and officially support), and ignoring the others. Complicated software is usually fragile enough that it will have similar results.

>Almost always, nobody has any intention to "make it run buggy."

Well, Microsoft has a long history of "making things buggy" everywhere else to create lock-ins. This instance may not be strictly related, but people's attitude has been shaped by years of anti-competitive stuff that MS has pulled.

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#282

Post by sah2ed »

dragonwriter wrote: post #220
mamon wrote: post #216

Earlier quoted context omitted.

For one, every Linux distribution comes with either Firefox or Chromium preinstalled. Is that monopolistic behaviour? Should EU require Linux developer to provide "Choose your web browser" wizard popping up on every new Linux install, similar way it is on Windows?

Microsoft's​ behavior was illegal not because of bundling alone, but because of bundling a product (IE) in a different market with a product which was a monopoly (Windows) in what was held to be (based on additional Al evidence beyond mere bundling) an attempt to leverage a monopoly in one market to monopolize a different market. So, no, it's not like any Linux distribution bundling a third-party browser like Firefox…

Honest question: if Microsoft didn't bundle IE and the user hadn't received a copy of those ubiquitous AOL CD-ROMs which came in the mail/magazines (and sometimes carried the Netscape Navigator installer), how is the user expected to consume content on the web without a web browser?

Surely, you'll at least need a decent web browser to get online to download a different browser of your choice, no?

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#283

Post by JshWright »

Lievelingsduif wrote: post #275
erikpukinskis wrote: post #251

Earlier quoted context omitted.

> If a user agent change makes the software slower for a platform then that's not exactly an accident so somebody must have had the intention to make it run buggy That's not necessarily true. For example they might be doing polyfills that are slower than the native windows implementation. I feel like I'm seeing a lot of this kind of thinking everywhere. The canonical example for me is lately is "he voted for Trump/Hi…

I'm guessing you don't really understand what he's saying. He's saying that by pretending to be Windows instead of Linux the webapplication runs better. There's nothing native about it. The same exact website/piece of software runs better by saying "Hi I'm on windows". This obviously means foul play.

> This obviously means foul play.

No, it really doesn't. It could just as likely mean they load a bunch of slow polyfills and workarounds when the browser is not part of a whitelist (assuming it won't have support for various APIs that it actually does support).

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#284

Post by endisukaj »

sdenton4 wrote: post #194
endisukaj wrote: post #156

Earlier quoted context omitted.

I've been using Linux for the past 6 years (on and off only the first two of those) and I have never had this happen to me. Maybe when I was using Ubuntu it was harder to find good answers that were not specific to someone else's usecase because of the nature of Ubuntu forums.

Back in the day I chose Gentoo Linux entirely based on an evaluation of its forums. It didn't disappoint: In five or so years of heavy use, I only ran into one issue that wasn't explained in great detail in the forums. The ridiculously high bar to entry seemed to do wonders for discussion quality.

I agree. I had the same experience as you when I moved from Ubuntu to Arch, the forums and Wiki are of extremely high quality.

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#285

Post by derefr »

sqeaky wrote: post #150
derefr wrote: post #87

Earlier quoted context omitted.

UA "detection" is an unworkable mess (i.e. a bad idea) when it's trying to differentiate between different consumer web browsers. But UA detection works just fine when it's only trying to differentiate between browsers as a whole, and things that are not browsers—like web spiders, or curl(1), or your native client app. I have a feeling that this is the latter case. The detection code was likely written to give a high…

Doesn't giving web crawlers different pages than the user facing UI get you kicked out of search engine indexes? The original ticket indicated different pages were served, not just QOS.

(I think) we're talking about pages served to logged-in users—i.e. the pages for files and folders in a given user's OneDrive account. Web crawlers wouldn't be able to reach those either way, so it (inconveniently from a fail-early perspective) wouldn't affect anything if the detection code does something stupid/crazy in the case of crawlers.

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#286

Post by neuronsguy »

Lievelingsduif wrote: post #275
erikpukinskis wrote: post #251

Earlier quoted context omitted.

> If a user agent change makes the software slower for a platform then that's not exactly an accident so somebody must have had the intention to make it run buggy That's not necessarily true. For example they might be doing polyfills that are slower than the native windows implementation. I feel like I'm seeing a lot of this kind of thinking everywhere. The canonical example for me is lately is "he voted for Trump/Hi…

I'm guessing you don't really understand what he's saying. He's saying that by pretending to be Windows instead of Linux the webapplication runs better. There's nothing native about it. The same exact website/piece of software runs better by saying "Hi I'm on windows". This obviously means foul play.

No. I think he understands exactly what the article means, and is correctly stating that it is not necessarily foul play at all.

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#287

Post by toqueteos »

solomatov wrote: post #175
cvrajeesh wrote: post #130

my 1$ hypothesis - partner company to which Onedrive UI project outsourced didn't had any Linux machine and all developers/QA working for this partner company were Windows users, to improve compatibility matrix they had kept few Mac machines in the office to run tests. Moreover this team might not received proper spec about the backend API at the initial stages of the project development, would have receive only at l…

I am pretty sure that they don't outsource development of such important functionality.

You'd be surprised of what things management chooses to outsource and later on insources because it's a steaming pile of shit.

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#288

Post by xtracto »

pjmlp wrote: post #169
hashkb wrote: post #151

Earlier quoted context omitted.

Just MS up to its old (classic) tricks.

Doesn't Hangouts only work properly on Chrome?

It is funny but in our experience around the company some people found out it works better on Safari than in Chrome (like, in Chrome oftentimes mic cannot be heard when one enters a hangout).

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#289

Post by bubblethink »

Lievelingsduif wrote: post #275
erikpukinskis wrote: post #251

Earlier quoted context omitted.

> If a user agent change makes the software slower for a platform then that's not exactly an accident so somebody must have had the intention to make it run buggy That's not necessarily true. For example they might be doing polyfills that are slower than the native windows implementation. I feel like I'm seeing a lot of this kind of thinking everywhere. The canonical example for me is lately is "he voted for Trump/Hi…

I'm guessing you don't really understand what he's saying. He's saying that by pretending to be Windows instead of Linux the webapplication runs better. There's nothing native about it. The same exact website/piece of software runs better by saying "Hi I'm on windows". This obviously means foul play.

No it doesn't. It is just shoddy detection. It may rely on some windows native APIs that may or may not exist on all linux variants. The fact that it worked by changing the user agents in some case doesn't mean it would be wise for them to deploy it for everyone. The proper fix would obviously be to actual detection and not rely on the user agent string.

Re: Onedrive is slow on Linux but fast with a “Windows” user-agent (2016)

#290

Post by princekolt »

rsweeney21 wrote: post #185

"Don't assume bad intentions over neglect and misunderstanding." - Hanlon's razor I worked as a Microsoft employee on the xbox.com website from 2007-2010. We didn't officially support any browsers on Linux because it represented We supplied a list of officially supported browsers to the customer support team. Any bugs on non-supported browsers would never get reported back to our team, so we'd never hear about them.…

Not supporting a platform is completely fine. Purposefully hindering performance when it is proven that the software runs well on said platform is one of the shittiest moves a company could make. I wonder to which other products they do the same thing, especially on places where evaluating such behavior is not as simple as changing the contents of a string.
Post reply on HN