Forgot your password?
typodupeerror

Comment: Re:Listen to Sales - as hard as it may be (Score 1) 159

by obarel (#48014977) Attached to: Ask Slashdot: Software Issue Tracking Transparency - Good Or Bad?

I guess you only buy bug-free software, then.

But seriously, isn't it better to see what's wrong and ask how the worse-looking risks are mitigated?
I guess the answer is about the general business culture, i.e. whether you're more likely to lose your job when the shit hits the fan if you say "I made an information-based decision and unfortunately the risk materialised" than if you say "I know nuffin'... they said there were no bugs".

Personally I'd get rid of a buyer who gave me the second answer, but I that's just me.

Comment: Re:First movers nothing. (Score 3, Funny) 108

by obarel (#47864399) Attached to: Why Google Is Pushing For a Web Free of SHA-1

There's no point in acting all surprised about it. All the planning charts and demolition orders have been on display at your local planning department in Alpha Centauri for fifty of your Earth years so you've had plenty of time to lodge any formal complaints and its far too late to start making a fuss about it now.

Comment: Re: This is why encryption isn't popular (Score 3, Insightful) 399

by obarel (#44529727) Attached to: Ask Slashdot: How Do I Request Someone To Send Me a Public Key?

Whether or not you want to trust a card given by the government is one thing.

But if the government actively encourages people to encrypt stuff then there is greater awareness of privacy and encryption. It means that more people understand the concept of private/public keys and are more likely to generate their own keys and use them. They're also no afraid of encryption as a concept (and a question such as "how do I ask for their public key without sounding like a geek" doesn't exist). I think that's a positive thing.

Other countries actively discourage privacy - yes, you can encrypt stuff, but if you don't give us the password then you'll end up in jail and we don't have to prove a thing. And why teach the masses to encrypt? It's so much easier listening to communication in the clear, and we can even perpetuate the notion that if you encrypt your files or communication then you're clearly hiding something and you're probably a dangerous criminal/terrorist/paedophile, because normal people don't use encryption.

Comment: Re: Have u thought about.. (Score 1) 524

by obarel (#43796507) Attached to: Ask Slashdot: Moving From Contract Developers To Hiring One In-House?

I'm wondering about the mainframe comment: does it mean that bug-free software was only ever written for mainframes? Because I'm absolutely 100% convinced that not all the software that runs on mainframes is bug-free, having worked on mainframes. Even IBM COBOL compilers had (have?) bugs, and they must have been tested by thousands of businesses.

Comment: Re: Have u thought about.. (Score 1) 524

by obarel (#43796413) Attached to: Ask Slashdot: Moving From Contract Developers To Hiring One In-House?

The reason is that for software to be perfect it has to be either proven or checked against every possible input.

The first may be possible in some cases, but is very time consuming for anything other than trivial exercises (and almost no one is willing to pay for that).
The second is simply impossible and some "representative subset" must be used for testing. This means that the once-in-a-lifetime case could be missed.

The car analogy doesn't work here.

Comment: Re:But...Agile teaches us... (Score 1) 400

by obarel (#43528811) Attached to: Dropcam CEO's Beef With Brogramming and Free Dinners

So what happens now with last minute customer requirement changes? I'm really trying to understand this (and hopefully learn from this).
If you have to push them in anyway, then you're still working to crazy deadlines.
And if you don't have to push them in, couldn't you have said "no" before moving to Agile? How did moving to Agile change your relationship with the customer?

Comment: Re:But...Agile teaches us... (Score 2) 400

by obarel (#43528721) Attached to: Dropcam CEO's Beef With Brogramming and Free Dinners

That's one possibility. Another is that whatever is supposed to be The One True Agile (tm) requires certain pre-conditions that aren't always met.

I could say "don't blame the single-pass waterfall process - if it failed for you, then you're doing it wrong". In some (rare) cases, single-pass waterfall is exactly right - a single programmer implementing a rigid specification (for example writing an H.264 decoder). But that's a pre-condition. It won't always fail and it won't always work, just like "Agile" or any other "methodology".

The truth is that there's no silver bullet. Every set of guidelines also includes a set of conditions (implicit or explicit). For example, most software development processes assume that the programmers involved are not all back-stabbing psychos. But even that's not always the case. Blaming reality for the failure of a process is the wrong way around.

If it's not in the computer, it doesn't exist.

Working...