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

The Parley Letter

david.heinemeierhansson.com

Post reply on HN
Print view View original

21–30 of 50 posts

Re: The Parley Letter

#21

Post by T-hawk »

unimpressive wrote: post #6
jeremymcanally wrote: post #4

Earlier quoted context omitted.

My guess is that iOS autocorrect strikes again. :)

Autocorrect is a bad idea. The kind of spelling error that iOS corrects (At least as far as I know.) is a syntactic error. Whereas the kind of error that autocorrect introduces is a semantic error. A semantic error is much worse than a syntactic one. At worst, a syntactic error leaves your meaning ambiguous. At worst, a semantic error leaves your meaning incorrect or inverted. Usually you can read past a syntactic er…

Well, it's a question of total value, weighted by frequency. If the autocorrection gets ten syntactic errors fixed for every one semantic error it introduces, it's probably a net gain for the writer.

Re: The Parley Letter

#22

Post by aGHz »

aGHz wrote: post #12

1970: " Another part of the necessary art of calling bullshit is to take down good ideas when they become inflated beyond what they can bear. Organizing your instructions in a clear way is a reasonable idea that I support. But there's just not enough there there to dress it up in function calls, machine optimization, and all the other -- here's that word again! -- wankery that passes for muster in the compilers ortho…

I should probably mention, before it's pointed out, that I mean REST is still in its infancy for web apps. It's more than proven itself as an excellent system architecture through the www.

Re: The Parley Letter

#23

Post by unimpressive »

T-hawk wrote: post #21
unimpressive wrote: post #6

Earlier quoted context omitted.

Autocorrect is a bad idea. The kind of spelling error that iOS corrects (At least as far as I know.) is a syntactic error. Whereas the kind of error that autocorrect introduces is a semantic error. A semantic error is much worse than a syntactic one. At worst, a syntactic error leaves your meaning ambiguous. At worst, a semantic error leaves your meaning incorrect or inverted. Usually you can read past a syntactic er…

Well, it's a question of total value, weighted by frequency. If the autocorrection gets ten syntactic errors fixed for every one semantic error it introduces, it's probably a net gain for the writer.

One abnormally bad semantic error can be worth a thousand syntactic errors.

Re: The Parley Letter

#24

Post by tokipin »

aGHz wrote: post #20
dhh wrote: post #17

Earlier quoted context omitted.

Let me translate it for you, then. Wankery: Complexity brought on by the love of an abstract idea without meaningful practical benefit. By definition I don't see the point in architectural exploits that can be labelled wankery.

I know, you've repeated that several times. "Without meaningful practical benefit" is what I'm referring to when I say you keep stating that you're missing the point. Just because you happen to not see the meaningful practical benefit doesn't mean there isn't one.

well the key word is "meaningful". and i think it can be stated in a more nuanced form:

wankery: architectural shenanigans which suffer significant diminishing returns on investment

Re: The Parley Letter

#25

Post by cunac »

pbiggar wrote: post #19
xentronium wrote: post #16

Earlier quoted context omitted.

To DHH credit, he always has fair points in his opinions, but the way he expresses these points is never diplomatic. For example, he couldn't resist making some blanket statements about java world and rspec. I am not sure about how intentional that is, but it surely is dramatical in 'flame war eliciting' sense of word.

The JAva point I felt was truly fair here. Servlets were good, but then it became the huge EJB which was bad. His rspec opinion - as expressed here that is - was "I dont like the DSL". Those seem fine really.

Just as a pointer I would recommend to get up to date with latest JEE specs and standards before making judgement What I see is most people base their opinion on EJB2 from past and they keep ranting about that. Also java ecosystem doesn't start and end with JEE

Re: The Parley Letter

#26

Post by andrewflnr »

unimpressive wrote: post #23
T-hawk wrote: post #21

Earlier quoted context omitted.

Well, it's a question of total value, weighted by frequency. If the autocorrection gets ten syntactic errors fixed for every one semantic error it introduces, it's probably a net gain for the writer.

One abnormally bad semantic error can be worth a thousand syntactic errors.

Have a little care, then. You have to be careful anyway with a touch keyboard. Autocorrect makes me faster, by correcting most errors without making me stop, and that's all I ask of it. I can't recall a time it's stung me in a major way; I can't count the number of times it helps me every day.

Re: The Parley Letter

#27

Post by cunac »

jrabone wrote: post #10
typicalrunt wrote: post #7

I think some clarity is in order. While Java certainly isn't a pinnacle of language design, it took quite a few years of devolution to go from something relatively simple like servlets to the steaming pile of shit that is J2EE I believe he means to say EJB here, and not J2EE, because Servlets are a part of J2EE.

Yeah, and I don't think he's been keeping up with Java EE 6 - EJBs are dead, long live JPA annotations. Of course, Spring is likely on borrowed time too. Just maybe we can lose the whole concept of a "framework", lightweight or not, that has kept those architects in business for so long.

why antagonism to architects ? It is only developer which actually will think about implications of what is doing from more than one point. It would be nice if I see more developers doing that instead just coding big pile of ..(you got a picture I guess)

Re: The Parley Letter

#28

Post by georgemcbay »

I agree wholeheartedly with with dhh's stated philosophy on the fact that code talks, bullshit walks and writing "future proof" code is a terrible idea (one that virtually every journeyman developer stupidly clings to for a while at some point in their career). But at the same time I'm not a Ruby fan, in large part due to the monkey patching ability he holds so dear.

Given the same philosophical outlook but perhaps less inherent trust in developers (including myself!) to always do the Right Thing, I've come to really appreciate Go as a no-bullshit language that does allow you to hang yourself (eg. the unsafe package, ability to ignore error returns via '_' variables, etc) but it at least makes you tie the rope to the ceiling before you can use it, so you have time to think about whether or not you really want to use it.

Re: The Parley Letter

#29

Post by mhw »

dhh wrote: post #18
typicalrunt wrote: post #7

I think some clarity is in order. While Java certainly isn't a pinnacle of language design, it took quite a few years of devolution to go from something relatively simple like servlets to the steaming pile of shit that is J2EE I believe he means to say EJB here, and not J2EE, because Servlets are a part of J2EE.

J2EE's sins include far more than just EJB. See http://en.wikipedia.org/wiki/Java_Platform,_Enterprise_Editi... . Servlets got just started running with a bad crowd, even though deep down it was a good kid.

Personally I wouldn't let servlets off that easily. One of the basic design choices they made was one request handling function per class (the service method, delegating to doGet/doPost/etc for http requests). As a result, most applications actually needed another framework on top of servlets to do routing based on request URL and mapping on to some model of server side components, like MVC. See for example struts, spring, webwork and a host of other java frameworks that were around at the time Rails appeared on the scene. At the time, Rails' routes file and having classes grouping related controller methods together was a clear demonstration of what the servlet API had got wrong.

Heck, the architects even designed it as a completely generic request/response framework with http as just one possible protocol. When http is your 99.9% use case, that just smacks of over-engineering.

Re: The Parley Letter

#30

Post by typicalrunt »

dhh wrote: post #18
typicalrunt wrote: post #7

I think some clarity is in order. While Java certainly isn't a pinnacle of language design, it took quite a few years of devolution to go from something relatively simple like servlets to the steaming pile of shit that is J2EE I believe he means to say EJB here, and not J2EE, because Servlets are a part of J2EE.

J2EE's sins include far more than just EJB. See http://en.wikipedia.org/wiki/Java_Platform,_Enterprise_Editi... . Servlets got just started running with a bad crowd, even though deep down it was a good kid.

My mistake, you were intentionally casting a wide net. I see where you are coming from now.

In my time working with Java (14 yrs now...sigh) for simple Web applications, I've never found the need for the majority of the J2EE stack, instead opting for Servlets and...well, just normal Java code. I've had Java gurus throw proverbial books at me for not using EJB and its ilk, but I just don't like how complex it all gets once the full J2EE stack is included.

Post reply on HN