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

Apple's Mistake

paulgraham.com

Post reply on HN
Print view View original

231–240 of 285 posts

Re: Apple's Mistake

#231

Post by rbanffy »

khangtoh wrote: post #213
eelco wrote: post #211

Earlier quoted context omitted.

No, just the symbol table in the binaries. You can easily do it yourself with the 'nm' command. It seems that they are using a tool that matches the 'nm' output with a list of private APIs since about a month or so, meaning that a lot of updates suddenly get rejected like this. What really annoys me as a developer though, is that they should be able to automate this check and give feedback instantly. (It's not like t…

Seriously if they are so hung up on undocumented API calls, they should just implement that check as part of the binary upload process. Run the binary through an automated flagging system. This will not only benefit developers but also cut their review time plus reduce the number of update submission indirectly.

This lack seems a symptom of what pg is describing - they are no longer getting the best and brightest programmers around.

Re: Apple's Mistake

#232

Post by rbanffy »

emmett wrote: post #95
brown9-2 wrote: post #60

Earlier quoted context omitted.

This might be a controversial opinion, but does anyone else think that if the QA team is reading your source code then you are doing QA wrong? Or do you really mean "held up by code review" when you say "get through QA"?

No, at some companies people really do have the QA department ensure "code quality". After all, it's Quality Assurance, so they must know about code Quality!

The best QA teams have developers and security people in them exactly for this reason. Quality is not what the users see - it's a correct implementation of what the specs say, built upon a series of basic assumptions that apply to all software they ship.

Treating QA as a step after development is a mistake that causes endless suffering to users.

Re: Apple's Mistake

#233

Post by st-keller »

Paul's new RFS "development 4 handhelds" seems to be quite simple: The only problem to be solved is: How to design a IDE that doesn't need a keyboard. I admit that this may be a bit diffcult as long programs are "written". Does anyone have an idea, how programs can be "sketched" or "told"?

Re: Apple's Mistake

#234

Post by ggruschow »

lucifer wrote: post #140

Paul, when you wrote this, did you mean by "programmer" anyone who started programming in the new century? Because Dave Winer -- and here I'm feeling like you did when you bought the iMac ;) -- wrote extensively on Apple's treatment of developers in the 90s. This is not news.

I don't think he was referring to programmers from any period. I'd be surprised if work for Apple computers ever constituted even 25% of programming work, and I'd be shocked if Dave Winer's reach was even 1%. I don't mean any offense by either, it's just it's a really big world.

Re: Apple's Mistake

#235

Post by rbanffy »

bad_user wrote: post #215
petewarden wrote: post #35

Apple has always treated third-party developers with suspicion. The developer website is only accessible by creating an account and agreeing to their license agreement. They churn through and deprecate APIs comparatively fast. They will quite happily steam-roller over third-party developers by adding competing features to the OS: http://en.wikipedia.org/wiki/Dashboard_%28software%29#Compar... None of this is new, the…

> The trouble with this comparative openness is that you end up with 800 pieces of software written by different companies fighting it out on your desktop, leading to crashes and inconsistent UIs all over the place. And that's a bad thing because? Who has the mainstream OS now with over 87% market-share? Even popular cross-platform applications that were first available for Macs (like Adobe Photoshop) are now more op…

"you can install Windows on any hardware you want as long as it's compatible"

Sorry, I had to laugh.

Part of the blame for Itanium losing the 64-bit bandwagon could be pointed to Microsoft, for not having Windows ready for it when it became available. Or for delaying the launch - it's silly to launch a processor without an OS to run it on.

Having Windows support for your hardware is paramount to any desktop computer company since the early 90's (with the notable exception of Apple). Alpha, MIPS and PowerPC (as in PReP) desktop systems are not dead because x86 systems were a better price/performance choice: they are dead because Microsoft pulled the plug on NT for them (and Linux was not ready to take its place at that time). That's also why all our computers are remarkably alike from the inside. It's not that Microsoft is open - it's that they have computer makers cornered.

Today I can get a non x86 computer and expect all my userland to work (with the possible exception of Flash, Skype and parts of Eclipse). I could not do that in the 90s.

The Itanium is not an expensive server chip because Intel cannot build an Atom-like processor with the Itanium ISA (heck - it must be easier than do it with the x86 ISA). It's an expensive server chip because it won't run Windows 7 and Office.

I see very little competition in the development tools in the Microsoft space these days. Tool makers know better - if they make something that competes with Visual Studio, Microsoft will come after them in the next release cycle and it is not worth the pain to fight them. That's why every non-Microsoft development tool around is built for a niche. Or is free, a case where Microsoft will not be able to go after you. That could be a reason for the free language explosion we are seeing.

And thanks. The Psystar imbroglio is a statement of how evil Apple can be about competition these days.

It even seems that after stopping attracting the best and brightest, they started attracting former Microsoft minions...

Re: Apple's Mistake

#236

Post by cesare »

c1sc0 wrote: post #212

I cross-post this from the other thread I started since I noticed the Apple heat is in this one, if you want to learn the background about my app, check this thread: http://news.ycombinator.com/item?id=951360 Below is what I've learned after 2 months of AppStore dev. Here are a few dirty tricks & ideas that I hope will make life in AppStore purgatory easier: Use the 'Easter Egg' field to plead directly to the reviewe…

> I got an app rejected because the reviewer did not like the price I set.

That's a new one. At least for me.

Re: Apple's Mistake

#237

Post by rbanffy »

YuriNiyazov wrote: post #45
onewland wrote: post #32

Earlier quoted context omitted.

In high school, I thought that the CodeWarrior IDE (which was, as far as I knew, the dominant development environment) was absolutely terrible. I'm not sure how good my judgment at that time was. That said, with OS X they really became a lot more developer friendly just because they adopted a UNIX base. At this point, I'd say they're the default platform for ruby/web development. Look at all the rails job postings fo…

The relevant times under discussion are: the era of the Apple II which came with built-in Basic, and the era of post-OSX but pre-iPhone

The Apple II... A computer you could turn on and start programming in 3 seconds... When will we be able to do that again ;-) ?

Re: Apple's Mistake

#238

Post by maurycy »

The sad fact is a complete lack of alternative. Personally I hate Apple so much. However, I still buy one or two Apple computers a year.

If you're a developer, and you want a stable environment that gives you access to the Terminal (bye, Windows!), has nice typography (bye, Linux!), excellent text editor (TextMate is the best one I found for Ruby) and does not cause stupid problems with printers or wifi (bye, bye, Linux!), you have no choice but OS X.

Their hardware is a piece of crap. Even a company that made an ad with them, agrees with that:

http://37signals.com/svn/posts/1489-every-mac-ive-owned-has-...

I think it's very similar if you are a professional photographer. Windows is so bloated it's painful to work with. Linux has no Adobe toolkits and you end up with OS X again.

Re: Apple's Mistake

#239

Post by rbanffy »

nickpp wrote: post #72

Laptops displaced desktops?! You are deluded. We develop on 8GB i7s with dual 24" screens. We have one at home and one at work, Remote Desktop-ing from each other. Gigabit ethernet and optical fiber internet. No molasses wifi here. We work FAST, we work HARD. But we use Windows 7 too, so we're not limited to Apple's crappy hardware either...

Ever since I moved from Windows to Linux I never had the need to develop on such a monster. My netbook (with some extra memory) serves me well. When I am at my desk, I just hook it to the big monitor, a keyboard and mouse and away I go.

The only moment I wish for an i7 is when I have to buildout Plone, which is time consuming, but specially so on a machine designed for battery life. Fortunately, I have been doing mostly Django work, so I haven't been building Plone out lately.

I wonder if other Linux developers also feel the need for 8GB i7 machines.

Re: Apple's Mistake

#240

Post by rbanffy »

No match for "WOMANWITHHAMMER.COM".

>>> Last update of whois database: Fri, 20 Nov 2009 12:10:22 UTC Anyone wants to give it a try?

Post reply on HN