HNHacker News
TopNewBestAskShowJobs

borland

297 karma · joined August 4, 2010

submissionscomments
borland··on Apple’s refusal to support Progressive Web Apps is a detriment to the web
Full disclaimer, Swift is my favourite language by a long shot and I'd love to use it everywhere, but it really is very complicated in some places.

Consider, for example, the interaction between generics in structs and classes, and protocols with associated types, and why you have to make stupid type-erasing wrappers like 'AnyObject' and so forth.

JS barely even has types at all, let alone generics. There's a lot less to go wrong there :-)

borland··on Apple’s refusal to support Progressive Web Apps is a detriment to the web
Very much agree with this.

Having developed many websites and many apps, HTML and CSS are really quite terrible when compared to something like iOS' auto-layout system, or even something like a DockPanel in Xaml (WPF on windows)

However, you wouldn't want to layout a document in Xaml or auto-layout either. Different strengths for different original purposes.

Personally I feel like a non-trivial part of why "web apps" (i.e websites pretending to be apps) generally suck comes down to this impedance mismatch

borland··on Swift is like Kotlin
This is the exact opposite to ruby, which uses `0..n` as inclusive, and `0...n` as exclusive

http://stackoverflow.com/questions/9690801/difference-betwee...

I always remembered it as "the third dot makes it bigger so it pushes the range and the last item falls off". I can't remember where that came from, perhaps related to the poignant guide:

http://poignant.guide/book/chapter-3.html#section2

I actually find in practice the swift ranges are the only ones I can reliably use without having to stop and look up the reference syntax every time.

For some reason, for me, `0..<n` reads as "0 through less than n" which signals to my brain a clear signal of an exclusive range.

Then I can work from there and if it doesn't have the < symbol it must be exclusive.

I suspect `0..=n` would work similarly well, but I've not seen a real language to ever do that yet

borland··on Finding an Alternative to Mac OS X
A plus one from me. I've used WSL for the last couple months to run a bunch of ruby and postgresql based stuff and once I got past a few teething problems upgrading it from ubuntu 14.04 to 16.10 it's been excellent. The performance of linux ruby under WSL is MUCH better than windows ruby. I'd recommend WSL to any ruby dev on windows simply for that.

A co-worker is also using it to run cross compilers for embedded linux hardware devices (various patched versions of GCC and such) and it's been flawless for him, which I found quite impressive too.

borland··on Ask HN: Labour of love projects
Urgh. I don't agree that Winmerge is "beautifully polished", even measured against the low bar of open source software. It's ugly and hard to use - I'd switch to beyond compare every time.
borland··on Ask HN: Java or .NET for a new big enterprise system?
Personally I'd say .NET, but I've got a windows development background.

On many angles (open-source project count, speed of most-optimized code, etc) I think Java is clearly ahead - however on many other angles, I think .NET wins.

C# is a much nicer language to program with, having all kinds of nice things like type inference, async/await, generics that work properly, proper closures, anonymous types, LINQ, and other things that Java doesn't have, and probably never will.

I also think the memory and disk footprint of a .NET server application is likely to be a lot lower than that of Java - In my company we have some REST api's written in C# running on windows, and some comparable REST api's written in java, and the C# ones consume about 100MB of RAM compared to over 1GB for the Java ones. The java programmers tell me this is a side effect of the open source project culture - A typical .NET service might pull in half a dozen third party libraries whereas a java service might pull in dozens.

.NET Core running on linux I find to be really interesting - it's just as fast as the windows one, if not faster, and uses a comparable amount of memory (I.E. still heaps less than java) - I'd go with that if I were starting today. Yes it's very new, but remember the core parts have been battle-tested in .NET for windows for over a decade, so it's not that new. Also if you're developing a big new enterprise system I'd guess your timelines are in the multi-year range. .NET core is going to have a long time to mature before you need it for production critical systems.

Looking further ahead, to be honest I like the look of .NET more than Java as well. C# 7 ships very soon with a bunch of nice language level features, and microsoft are really pushing forward with .NET core and making a lot of improvements. Java on the other hand is moving a lot more slowly, and since Oracle purchased sun there has been a lot of noise in the java OSS community about Oracle abandoning large parts of it and basically letting it fall to bitrot.

P.S. As much as I like Visual Studio, IntelliJ is a lot better... Just make sure you've got heaps of RAM :-)

borland··on Ask HN: How do you manage and structure a team?
> judging the direct merit of each others work is hard. Mostly because the work is a spectrum between “core” work, that is required, and “new features”

You can't implement the new features without the core work to provide a foundation. Don't judge performance by financial impact of the code, judge performance based on peer review. If you've got 2 programmers, you'll want them to be doing things like design reviews and code reviews, and as such they'll be able to provide a reasonable assessment of how well they think their peers are doing. Whether each bit of work itself needs doing in the first place? well you need a signoff/approval process. Whether that's a manager, or simply a chat with your peers - the act of convincing others that yes I do need to implement this core XYZ instead of feature Q is an important one

borland··on Apple Maps Connect
BEWARE. When I try log in it says my apple account has been locked out for security reasons. I went and reset the password and successfully logged into a bunch of other apple stuff, but when I tried to log into maps connect it said I was locked out again.
borland··on Catch: A modern, C++-native, header-only, framework for unit tests
I really like the look of the "Sections" for managing setup/teardown of tests. I'm envious and next time I'm hacking on our in-house C# test framework I'm going to try very hard to implement them! Neat!
borland··on iOS 7 is Yucky. Why Isn't Anyone Saying So?
I like iOS 7 a lot. Why isn't anyone saying so?
borland··on Poll: Do you use Google Reader on a daily basis?
Where's the "Yes, but I never actually use the interface, only third party apps like Reeder on iOS" ?
borland··on Collect HN: Aprils Fools
UTC+13, even. We're still on daylight savings time in NZ
borland··on Is it ok to use warez software at a startup?
Of course not! You're trying to sell software right?

If you had some customers at your startup coming to you and saying "Well, we want to use your software, but we don't like your pricing, so we're just going to pirate it instead, is that OK?"

Would you say "Sure, go ahead, we just charge a price for the fun of it"? If not, then you know that it's not OK. If yes, WTF?!?!

borland··on Apple deprecates Java
I get slowdowns caused by text rendering in terminal when rsyncing large numbers of files. It's easily worked around though, just minimize the terminal window
borland··on [dead]
I'd disagree with web workers needing to access the DOM. If you're storing application state in the DOM, you should reconsider the design (or lack thereof) of your application.

And sure, I'd like IE9 support too. And a pony.

← PreviousPage 3 of 3