Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Comment Re:Wrong! (Score 1) 214

Actually, it probably was written by the bank. Banks have IT departments, and they often write their own custom software.

Disclaimer: I work for a company that writes software for banks.

At least our clients all have IT staff that at least is able to modify database directly. They can't change the code we write, but if they need to make configuration or input data that we did not anticipate, whey have access to the database and can do the magic themselves. In this case overriding age input should be trivial, but perhaps bank was reluctant to change things without testing for side effects first.

Comment Re:So what's the issue? (Score 1) 214

I agree about manual update of the database, but in this case it seems that software patch was needed. It did surprise me a bit that bank didn't have admin to do it. One other solution for situations where input is improbable yet possible possible: application can ask the operator to check or re-enter the dubious input.

Comment Re:So what's the issue? (Score 4, Insightful) 214

There is no way to take every edge case into account when designing systems. She is clearly an edge case. Why is this even here?

Because when you think “this will work for 99% of cases” the corollary is “the rest of 1% can go fudge themselves”. Sometimes that is fine, but if a person can't get their pension, that is certainly not ok. Consider this a cautionary tale for programmers.

Comment Re:virtue signaling (Score 1) 478

While I admit that I don't like the SJW type (and I'm inclined to believe you are one),

You missed the point, there is no such thing as a SJW, it's just a way for you to show a tribal affiliation.

Your pathetic attempt to reinterpret my text, and in the process put in a few cheap shots on my person, suggests you are feeling threatened, which in turn indicates I came entirely too close to the truth for comfort.

Projection. You're not worth insulting. That was also the point: you're not saying anything, you're just raising the noise floor. I am also compelled to point out that in no sense were your words "reinterpreted", and I said nothing to suggest that your words should not be interpreted in their literal meaning. It would be convenient for you if that were true, and it would save you an awkward self-realization.

you're a wanker.

Deflect, dehumanize. How tedious

Comment virtue signaling (Score 4, Insightful) 478

That position is exactly as valid as saying that the opposition to "Social Justice" is merely a bunch of amoral recalcitrants.

But really you're just using "Social Justice" to mean "people I don't like". Because to the degree that that has anything to do with this subject, basically the strongest argument that can be brought to bear would be that the "slave" women have internalized the Patriarchy to the point of self-degradation. But if you're wont to hold that opinion, there are quite a few things higher on the list than private sex games.

What you're doing is virtue signaling. It's not very intelligent and rather boring.

Comment Re:Missed opportunity (Score 1) 191

Google missed an opportunity here. They should have programmed it to respond with something like "something almost but not completely unlike a hamburger" or gone with kickback money from McD's and said something like "a pale imitation of McDonald's quarter pounder" or even "hamburger royale".

Google was just quoting Wikipedia and it was swiftly edited:
https://en.wikipedia.org/w/ind...

Comment Re:More memory the longer it runs (Score 1, Flamebait) 213

"disable extensions" or "You're wrong"

Sounds about right. Why should Mozilla fix memory issues of other applications. If extension developer writes leaky code, what should FF do? Balcklist or restart extension somehow? And if you have 12 tabs of “web applications” like twitter, should FF reload page?

I'm not saying Firefox is without flaws or anything, but it is a sort of operating system on it's own. Keeping that in mind, we can paraphrase Maxo-Texas statement like this:

Right now I have 12 applications open and they're using 923 mb of memory and 2.7% of cpu (on an i7).

This is hardly scandalous. Without knowing details, just pointing out MBs and tab count is useless

Comment Re:Learning the same lesson over and over (Score 1) 104

Why are you taking this simpering symcophanitic line that is wildly divergent from reality?

It seems that we are in a misunderstanding. I am not trying to suck up to you. I have no idea who you are and I'm sure this relation is mutual. And I'm pretty sure Lennart isn't reading any of this either.

The guy is no genius and hasn't even been coding for as long as this site has been up.

I never said he is genius or that he is any good at what he does. But apparently he does not share the same vision as you or the loudest /. community members. Please note that having “vision” is not a sign of any kind intellect either.

Seriously, read his blog.

I did read his blog when systemd was the new kid in town. But I don't see how it has to do with my speculations about how people here on /. perceive systemd.

Comment Re:Learning the same lesson over and over (Score 1) 104

He's trying a very different approach to the earlier init systems and it shows

Indeed it is true. Lennert sees chaos and systemd as a cure. You might see a beautiful diversity and modularity, which the systemd stomps on with the one true way. And both of these views have merit, and both have tradeoffs. Take Linux kernel. It is a monolith and you can't easily swap one subsystem for another and be sure that next version won't brake it. It sucks for some, but the community thinks it is better for Linux. I don't know which is the right call here.

An init system should not be, especially one developed with an aim to run some parts in parallel (obviously not all since it still hangs in some situations where the earlier init (correclty) gave up, reported an error and let the next task run).

At least in theory systemd should kill hung process (of course, if it was crucial for system, then boot will fail). The parallelism is not like “all processes at once”, but dependency based. If you call process P that requires Q and R, systemd will start Q and R in parallel (if they are not dependent among themselves). If the configuration is faulty and Q and R are interdependent, then bug is in the configuration. This can happen in classical init, where by wrong configuration startup sequences will fail (like starting DHCP client before networking).

He knows MS Windows and his development environment as distinct from a *nix production environment.

Not sure about this one. IIRC systemd was inspired by launchd on MacOS. In any case, I don't believe that all *nix idioms are always the right solution. They do hold up pretty well, but when a new solution is proposed that breaks *nix conventions, it's not automatically bad just because it's not *nix way.

Comment Re:Learning the same lesson over and over (Score 1) 104

Nobody else was paying for as many developers so all the other distros are repackaging RedHat's work.

I agree that RedHat investment helped to develop systemd and that RedHat can force systemd on RHEL and Fedora, leaving it's forks with little choice. But I don't believe that Debian, or Arch or openSUSE adopted it just because Red Hat did. If it was the case, rpm based package management would be on all Linux distributions.

There's still a design of binary logging, and not much of it, with a race condition that would earn a fail in a school project. There's been a stupid move to disable background tasks when a user logs off.

Let me guess, you have read countless discussions on slashdot and elsewhere about how binary logging is awful, then someone points out that plain-text logging is still there live and well. But the only thing you remember from the conversation is “binary logging”.

As for killing user background processes, it's a change of default option. You can change this default with compile flag or in /etc/systemd/logind.conf. In other words, it's up to the distributions to do what they feel is right (and for example, Fedora 25 does not kill by default), and if admin feels that distribution made the wrong choice, they can change one flag. I guess that after a month the only thing you will remember about background processes is that systemd kills them.

As for race conditions, I just don't know. Software as such is prone to such bugs, especially ones written in C, but given that your other examples of shortcomings of systemd where not that accurate, I'm not sure even this is not just some urban legend.

Slashdot Top Deals

One good reason why computers can do more work than people is that they never have to stop and answer the phone.

Working...