Re: Armory crashing firefox consistently now

Pla <pla@xxxxxx> wrote:
No it doesn't. No software can be made to be %100 secure, it's
impossible. You don't even know where the problem resides.

So you don't even try?

Do you know what an asymptote is? That's the curve that measures software
quality versus expended effort debugging it. Sometimes called the 90/10
rule, where "The last ten percent of a project requires ninety percent of
the effort". Software debugging is the same way. Sure, 100% reliability
and security may not be possible. But you don't let flaws pass ignored.
That's called laziness and negligence.

You're ("you are") wasting ridiculous amounts of effort defending your
softwares' right to be crap. And the authors of said software wouldn't
even agree with you!

Go ahead, try to repeat what you've said here to someone like Linus or
Theo de Raat. Just tell me where to find the posts, because that thread
would be short and very hilarious. Especially in Theo's case.

One can very easily do things in java for instance that will cause
functionality problems, that's not the fault of java, hell I could do
that in any language, once you exicute it problems can start. Improper

Java the language or Java the JVM?

If you can exploit and attack the sandbox model in a JVM, then publish the
exploit code to Bugtraq. Sun and the other JVM vendors will jump on it
like nobody's business... just like they have, in the past.

But that's not really relevant, because the source language is irrelevant.

If you can write bad code and run it yourself, then you've *become a software
publisher*, and you are responsible for the bugs in YOUR code! The difference
between you and someone who knows what they're doing is that *they* usually
want to fix the bugs, and will value bug reports.

If you're doing something as idiotic as a forkbomb, or malloc-ing all
available memory, the vendor will close the bug and refer you to the
existing resource controls built into the OS. Or, in the case of a JVM,
the max memory/heap command flags.

Did you have any other point besides saying, "Look, I can write bad
software, and I don't care?"

Besides the issue with the armory causes a temporary lock up of the
browser -- it is not 'causing an all out crash.

Then that's a starvation of resources issue, or a bad algorithm. Regardless,
you can still file a bug report on it. The only difference is the main tool
will be a profiler rather than a debugger.

--
Brandon Hume - hume -> BOFH.Ca, http://WWW.BOFH.Ca/
.

Relevant Pages

  • Re: Static vs Dynamic
    ... it's easy to miss the bug because the code looks quite correct ... >>As an experienced Java programmer you need to keep lots of code idioms ... These problems don't exist in a dynamically typed language. ... Any bad thread programming in any language is troublesome ...
    (comp.lang.lisp)
  • Re: Why Generics?
    ... > Every study I've ever heard of concludes that the cost of a bug ... ``Smalltalk v. Java: Which is More Productive? ... As essentially a rather static language, Java needed generics. ...
    (comp.lang.java.programmer)
  • Re: Why does Safari use so much processor power?
    ... Java in my daily work and think the language is quite good. ... File a bug at bugreport.apple.com about it. ... Lousy code is lousy code on all platforms, ...
    (comp.sys.mac.system)
  • Re: Why does Safari use so much processor power?
    ... Java in my daily work and think the language is quite good. ... File a bug at bugreport.apple.com about it. ... I am betting this because I never see a thread sleep in the system ...
    (comp.sys.mac.system)
  • Re: Groovy experience ?
    ... on a JVM rather cancelling each other's popularity out. ... things would have gone had there been only one language... ... As far as 'diversity', speaking for myself, with Hecl I am taking the approach, to some degree, that I want it to be different from Java. ...
    (comp.lang.java.programmer)