Jump to content

Wikipedia:Reference desk/Archives/Computing/2014 February 21

From Wikipedia, the free encyclopedia
Computing desk
< February 20 << Jan | February | Mar >> February 22 >
Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


February 21

[edit]

Uninstalling the $%^&ing Respondus Lockdown browser

[edit]

Does uninstalling the $%&*ing Respondus Lockdown software completely undo the wide-ranging system changes caused thereby? If not, how can I repair the security holes that it introduced into my system? Thanks in advance! 24.5.122.13 (talk) 06:57, 21 February 2014 (UTC)[reply]

I'm not familiar with the software you've mentioned but System_Restore might help, provided you have the feature turned on and you can find a recent Restore Point that is not going to create more problems than it fixes. It should undo any changes made to the registry. Obviously you need a restore point from before you installed the software.196.214.78.114 (talk) 10:12, 21 February 2014 (UTC)[reply]
I'm familiar with System Restore (in fact, I've had to do it once already on my current machine, and at least twice on older machines), but it sounds a bit drastic because it would probably also roll back other updates that have nothing to do with this piece of malware masquerading as a browser. Still, I'm willing to do it if need be, but I don't want to do it needlessly. So can anyone tell me if removing the Lockdown "browser" completely reverses the system changes that it has made? Because if it does, then there's no need to do anything else -- but if not, then System Restore (or even System Recovery if there's no suitable restore point) may well be in order. 24.5.122.13 (talk) 07:25, 22 February 2014 (UTC)[reply]
It might be too specialised a product for anyone here to know about, you may be better asking on their support forum or submitting a ticket[1]. It you do want to uninstall it, check things like your firewall and the registry. I've never tried the software but it looks like a browser with a very specific function, locking down the computer so you can't cheat in a test, I would not class it as malware as it is not malicious software. You may need to undo the things needed to get it to work, do the reverse of this page: turn on all virus checker, set security setting back to high, check firewall exceptions and the port block list, reenable any popup blockers. You may also want to uninstall java which it seems to relay on. If you really worried, run some hardcore malware scanners.--Salix alba (talk): 09:12, 22 February 2014 (UTC)[reply]
System restore is a fairly benign task, if you have only installed Respondus in the last week or month then restoring to just before it was installed is a good solution. Yes you have to rerun updates but for most programs thats a fairly automatic process.--Salix alba (talk): 09:12, 22 February 2014 (UTC)[reply]
Thanks! Just want you to know that I had already uninstalled the program before asking this question, and just wanted to know if I had to do anything else to completely restore my computer security. And I've scanned my system with two different scanners -- Norton says everything's fine, but McAffe says that my anti-spyware apps are disabled (for no apparent reason) and spyware definitions are out of date. Which should I believe? (Both are updated daily, BTW.) 24.5.122.13 (talk) 20:30, 22 February 2014 (UTC)[reply]
Why do you think it compromised your security? I couldn't find any online discussion of security holes associated with Respondus Lockdown, though I didn't look very hard.
I don't know what McAfee's scan meant when it said your anti-spyware apps are disabled. Is it scanning for third-party software like Ad-Aware? That would be weird, and I wouldn't trust it to do that reliably. If it's talking about McAfee products, it's most likely trying to scare you into paying for them. If you could quote the exact McAfee product you used and the exact message you got, I might be able to guess better. -- BenRG (talk) 21:51, 22 February 2014 (UTC)[reply]
All right then, so it sounds like my system is clean. (I'll re-scan it with Norton just in case, but it looks to me like the McAfee security scan is probably giving a false alarm because it doesn't recognize the Norton product.) Thanks! 24.5.122.13 (talk) 02:32, 23 February 2014 (UTC)[reply]

piecewise-linear optimization

[edit]

I have in mind a potential project that involves minimizing linear functions of a large number of variables with linear constraints — just the thing for the simplex algorithm, right? (Something I hadn't thought about in thirty years!) But wait, the function I want to minimize is not a simple linear function but the maximum of a family of linear functions. Does Numpy or Sage, for example, have a black-box simplex function that can handle this, or would I need to write my own? Or is some other approach more appropriate? —Tamfang (talk) 08:16, 21 February 2014 (UTC)[reply]

Around here, cvx is the standard. It's "sort of" open-source software. Nimur (talk) 08:56, 21 February 2014 (UTC)[reply]
It is possible to transform an LP problem with a piecewise linear objective function into a linear objective with extra variables and constraints, e.g., [2]. Programs such as AMPL can handle these problems--here is a book chapter on piecewise linear objective functions in AMPL. --Mark viking (talk) 19:52, 21 February 2014 (UTC)[reply]

Is html5 a programming language?

[edit]

According to wp, it's still a mark-up language, but since you can create html5 applications, and there are people talking about html5 programming or html5 developers, shouldn't we already call it a programming language? It is indeed one that run on a browser, but many languages are interpreted by something else, and are still programming languages. OsmanRF34 (talk) 19:09, 21 February 2014 (UTC)[reply]

If you do a search for an obvious seach term like 'html5 programming language' in a browser, whether or not your browser supports HTML5, you should find plenty of useful discussions like [3] [4] [5]. Maybe even [6] (not sure as it's Quora). All these sources in fact even the article you linked to emphasise that you need to consider there is a difference between HTML5 and HTML5 + CSS + JavaScript. Nil Einne (talk) 19:57, 21 February 2014 (UTC)[reply]
Pay attention, that doesn't enter into the implication of my question, of why it's not a programming language still. This discussion do not enter into what sort of algorithm it can not implement, or it's Turing completeness. Even if we don't consider it in combination with js. If I wanted a discussion I would have not come to the RF. So, don't point to irrelevant discussions. OsmanRF34 (talk) 21:33, 21 February 2014 (UTC)[reply]
I don't understand what you're trying to say, but HTML 5 (without Javascript, etc.) does not fit the usual notion of a programming language very well, especially because it doesn't have a computational semantics. It is a formal language, though (as are programming languages). -- BenRG (talk) 03:42, 22 February 2014 (UTC)[reply]
As BenRG has also said, you haven't really explained why you believe it is a programming language in any manner that I can really understand. Nor for that matter what you actually want. Your only statements you have made that I can make sense of and appear relevant are
1) Lots of programming languages are intepreted. This explains why it could be a programming language but it doesn't say it should be a programming language as I presume even you acknowledge plenty of intepreted things would not normally be considered programming languages. To use a simple example, all versions of HTML are clearly interpreted, are you saying they are all programing languages? Or if you are saying that, I don't get why you brought up HTML5 at all. (Note that a blog post linked in one of the sources I've provided does actually say that all versions HTML are programming languages [7] but it does give a far more sophisticated explaination then simply 'it's intepreted'. Which isn't surprising since I'm pretty sure you could come up with plenty of other things which are intepreted yet which even the blog post author wouldn't consider a programming language.)
2) People talk about programming HTML5 apps. Yet as the discussions I linked to and our own article explains, there is a big distiction between HTML5 itself and HTML5 + CSS + JavaScript. Most HTML5 applications would be nothing without at least one and probably both of these and maybe even more. You can do some fairly sophisticated things with HTML5 alone but most people do use various other additional stuff in their HTML5 applications. I.E. when people talk about programming HTML5 applications they are usually (although probably not always) talking about developing stuff using all of these and possibly more (for example server side programming). So the only real thesis of why HTML5 is a programming language you've provided has been more or less answered in the what I've already provided.
I don't get the relevance of much of your followup. Turing completeness is only mentioned by two commentators in one of the source, with no mentioned elsewhere as far as I can tell (I even checked the Quora one to make sure) and so is clearly not the main thesis for why HTML5 is not normally considered a programming language. The sources briefly discuss what can be implemented in HTML5 alone but it seems clear to me it's not the only reason why most of them suggest HTML5 is not normally considered a programming language, instead they generally go in to more sophisticated explainations (like BenRG did) about what they consider is needed from a programming language and why HTML5 lacks it. Although as I've said, ironically the only real thesis you've provided for why it is a programming language appears to be what can be implemented in HTML5.
Note that I used the term 'discussions' losely, one of the links is to a blog post on a university website which while probably not a WP:RS (the person who wrote it doesn't appear to be a notable subject matter expert) is not really a discussion either. But it does explain in some detail why HTML5 is not normally considered a programming language. (It also includes some discussion from commentators, including some who suggest HTML5 is a programming language which I presume you've read considering you mentioned Turing completeness and it doesn't seem to have been mentioned anywhere else. I'll be honest I didn't actually notice the discussion before this reply as my main point was the blog post which seemed decent enough.) Another source is potentially an RS but doesn't really directly speak on the question of whether HTML5 is a programming language. I presume you are already aware that in the case of the only real discussion source, as with many discussions, contain a number of explainations from partipants of why HTML5 is not normally considered a programming language, something which I checked before my first reply. (As I made clear, I didn't check the Quora one well but it does provide some brief explaination.)
If you look hard enough, you could probably find a RS that says HTML5 is not a programming language which I would note would be enough for any article dispute, unless you look even harder and perhaps find a RS which says HTML5 is a programming language and able to demonstrate that it isn't WP:UNDUE. But such sources won't necessarily explain why it isn't normally considered a programming language which I presume is what you want yet which the sources I have provided do attempt to answer that question (well except for one).
Perhaps you can find a RS that does both, but as I've said, it's not clear why this is necessary when you already have plenty of sources explaining why HTML5 is not normally considered a programming language, even if they are not RS. And you haven't actually given any good explaination that I can tell of what you're asking for that isn't answered in the sources provided which as I've said were easily found using a simple search.
I presume you aren't trying to edit an article, as you should be doing this in the article talk page then. (In any case I'm not sure there is any reason for any of this discussion in an article, calling HTML5 a markup language in our article is clearly the right thing considering thats what the spec itself calls it as the sources I provided said multiple times. And whether or not it should also be called a programming language doesn't seem sufficiently relevant since it's primarily irrelevant semantics or categorisation.)
Of course as the anonymous IP has said, as with many things, there's no strict definition of a 'programming language' so none of this means it's impossible to call HTML5 programming language (and from the sources I've provided there clearly are some people who do), simply that there are plenty of reasons why it many people don't call it one.
Nil Einne (talk) 04:56, 22 February 2014 (UTC)[reply]
I don't think HTML5 is interpreted, either, since that again implies some sort of execution semantics. -- BenRG (talk) 21:36, 22 February 2014 (UTC)[reply]
HTML5 contains several elements, one of which is a programming language. I guess (without having the polling data) that most people would not call HTML5 itself a programming language. But there are no strict mathematical/legal/etc definitions of these concepts. Is a spreadsheet a programming language - even when it is Turing complete? I suggest you should not expect or demand a definite yes/no answer. Or mu, or category mistake. 88.112.50.121 (talk) 22:02, 21 February 2014 (UTC)[reply]
If PostScript can be referred to as a programming language (as our article does in the first sentence), then I suppose html5 can too. Looie496 (talk) 16:47, 22 February 2014 (UTC)[reply]
PostScript is a programming language. It's Turing-complete (which, while probably neither necessary nor sufficient, is a pretty good indicator) and people have, at times, written real applications in it that e.g. used the Apple Laserwriter as a computational engine at a time when it's 68000 processor was beefier than many PC cpus. Of course, PostScript is mostly used for generating images, but then so is Logo with its Turtle graphics. HTML (including 5), as far as I know, is only a markup language (though again, the terms are not necessarily contradictory - LaTeX is both a mark-up language and a programming language). A programming language needs a way to describe reasonably general algorithms - I'd insist on at least functionality equivalent to primitive recursion, i.e. iteration over finite but conceptionally unlimited sets. JavaScript has this, but plain HTML5 does not. --Stephan Schulz (talk) 17:17, 22 February 2014 (UTC)[reply]