Putting that Shai LaBuff (spelling) kid in this one is probably what did it in.
Short Round was annoying, but nowhere near as irritating as a moronic woman who was screaming and being mocked by the other characters every 10 minutes.
Raiders was an awesome movie and it was one of the few action movies ever to have had a really strong supporting female character who could actually stand up for herself on her own merits, often challenging the main character. When Lucas and the studios tried to capitalise on the first movie's popularity, the Marion character was dropped and much of the really good stuff that made Raiders' was thrown out the window in favour of stupid 1980's movie cliches that were popular at the time. (Notably the annoying kid, and the annoying lady who can't look after herself and needs rescuing by more intelligent, stronger and more competent male characters, including the 10 year old annoying kid at times.)
It looks as if they've re-cast Karen Allen as Marion in the new film (even third on the IMDB list), so I'm hoping there might be something more than a cameo and that it's actually done well. I guess we'll have to wait and see.
Describe the sin function. Why the hell should it need an instanced class ? Takes a number, spit a number everytime with the same logic. No different than the - operator.
I'm not personally a proponent of OO-for-everything being practical today, but there are ways that languages can do things nicely if you're willing to think about it in a different mindset. OO is about thinking in terms of objects, not about spinning out annoying extra object instantiations all over the place. (Actually OO is about a lot of abstract and controversial things that go well beyond simply having objects, depending on who you ask.)
If I was trying to design a sin() function in an OO way, I'd probably put a.sin() method on the number class, to return another number object. If this idea is going to require too many thousands of mostly-redundant methods available on every Number object, then perhaps the language (which might allow for custom extensions of existing classes) would require that the programmer import a Math library of some kind before the.sin() method became available on those objects.
Only thing where MS dropped the ball, in my opinion of course, was that they dragged the abomination of VB to the.NET world. Now there's a heck load of VB.NET programmers who don't know a shit about object oriented programming and are doing everything in the old god forsaken ASP way.
I'm not sure they had much choice in all honesty. There are so many VB developers out there that there would have been an outcry if Microsoft hadn't given them some kind of clear path to move to.NET. Furthermore, many of those developers would have avoided moving to.NET at all, and would be writing awful code in just as awful programming frameworks that were no longer supported. At least Microsoft has cleaned up a lot of the worst VB quirks and inconsistencies when they moved it to VB.NET, such that it's now almost possible to write good code in VB.NET if you're so inclined. (Although why would you when C# is so much easier to write with once you understand it?). Bad code tends to be more predictable and do less damage on average.
Despite what's underneath Windows, programming it through the.NET platform is very slick. Most of what had to classically be linked to in obscure ways is wrapped in the Framework Class Library.
.NET is nice when you can stay inside.NET, but I've found this is very difficult unless you're developing a relatively simple application. As soon as you want to do anything fancy to interact with the underlying OS or other apps, it usually becomes necessary to start playing with COM wrappers and P/Invoke on at least some level. If you're lucky you might find a third party.NET wrapper library someone's written to do approximately what you want, but quality of those kind of things varies and it usually still requires some interaction with the less-safe APIs.
My experience trying to write a toolbar for use within our organisation would have been horrendous if I hadn't come across BandObjects (or the original), which is clearly popular if you look at the forum below, even though it's not actively maintained by its original author. I still came up with a variety of issues trying to get it to work in Vista, particularly involving getting popup menus that would overlap the Window taskbar properly, and detecting Windows colour themes in a nice way to make sure text was readable. Both of these things need some additional coding that bypasses.NET completely to talk directly to the Windows API using ugly hard-coded constants that mimic definitions in some anonymous Microsoft C++ header file (thus will probably lose compatibility sooner or later), and I still haven't gotten them to work nicely.
There does seem to be a certain irony in writing to Mr Zelnick's mother to complain about his upbringing when Jack Thompson's main argument, as far as I can gather, is that violent media, including video games, are primarily responsible for violence in young people and society.
If that's the case, why does he expect that she had anything to do with it? Surely she's as much a victim of a free state as all the other parents whose children are running wild and uncontrolled.
Instead of focusing energy on the ISO vote, focus on getting implementations of the standard that *you* think is reasonable into widespread usage. If you think it is ODF or RTF or HTML or any of the hundreds of file formats for document representation that should be the choice of governments, then get good, usable versions of software into the market.
I can think of at least two office suites (possibly three) off the top of my head which either use ODF as their primary document formats, or plan to in the near future. They've all done so because it makes sense for users to be saving in an interchangable and properly standardised format, and it means that somebody's application suite doesn't have to be decided by their archive of documents they have to be able to reliably open. Unfortunately the number of implementing office suites doesn't equate to widespread usage because the provider of the most popular Office suite, which controls the market and locks people in, has strong priorities that completely ignore what's best for customers.
Microsoft will never natively support ODF if they have anything to say about it, even if it is best for their customers. This is because their own unimplementable formats give it not just control over individual customers, but a virally* distributive control over everyone with whom their customers exchange documents. Consequently the most popular Office suite will always be decided by Microsoft's lock-in tactics rather than which suite is actually best for customers.
I'm sorry but if Microsoft's skewed priorities have led it to interfere with government processes and corrupt them at everyone else's expense, simply so it can keep control of the market and continue its monopoly, then I'm fully in favour of complaining and making it known as clearly and noisily as possible and every level where it occurred.
* If Microsoft can refer to the GPL as viral then I'm going to do the same for Office.
It's a shame he was billed for it and perhaps it wasn't unfair and should be considered that way, but any head injury is serious until proved otherwise, regardless of what the person who suffered it might think. It sounds as if the ski patrol people on-site were doing their job properly as far as looking after him. If you get knocked unconscious then you potentially have concussion, internal bleeding or swelling which can have non-immediate effects, and you need to get to a hospital as soon as possible to be checked out properly.
If I was out in a wilderness area and responsible for someone who was knocked unconscious, I'd definitely be getting them evacuated as soon as it was possible to make arrangements, irrespective of how quickly they woke up afterwards or what they said. If calling in a helicopter was the fastest way out, they'd get it whether they wanted it or not. It's basic first aid. It's true that sometimes this gets overlooked, especially for drunk people on the streets (to name an example), but it's especially important in any remote area where you could be a long way from help if you suddenly drop to the ground from an injury you didn't know you still had, because if that happens then you're likely to die.
But if it was a legitimate missing persons/accident, etc. as it appears to be in Fossett's case, then the next of kin should not be billed for the expense.
I'd presume that it's more of a claim on Steve Fossett's fortune rather than directly billing his widow, although I'm not a lawyer and I'm also not informed about whatever arrangements Steve Fossett had with the state of Nevada before he left.
As for taxpayer funding, I'm in mixed minds about whether people should be charged for their rescues. In New Zealand, and I'm sure many other places, we get lots of tourists who come here for the outdoor experience. Most of them are great, but there are still a lot of tourists who completely ignore all the advice about what outdoor safety gear they need and where they should and shouldn't go according to their experience, and consequently need to be rescued out of caves or off the side of mountains or from behind flooded rivers or wherever else they've gotten themselves stuck.
Rescue helicopters and their crews are expensive, and large volunteer search parties typically mean that people are giving up their daytime salaries to go and look for someone who got themselves into trouble. I can appreciate that sometimes things just go wrong, but if a person put themselves at risk through their own fault and triggered a rescue operation, I don't personally have as much of an issue in sending them the bill, or at least charging them an impressionable portion of the bill if it's unrealistic. (Sending a bill that will ruin someone for life isn't much help at all.)
Steve Fossett has a history of putting himself at obvious risk as part of his hobby. This is fine if it's what he wants to do, but when he triggers a massive rescue operation I can appreciate that some people might think he should be responsible for footing at least part of the bill, if not the whole thing. Whether that bill should be transferred to the person who's inheriting his vast fortune would, I think, depend on legal details.
In my experience, the number one problem with IT is that the programmers and managers really don't have enough interaction with the end users to understand their side of things.
I agree. I've been working in a medium sized organisation (~400 people, all in one building) for about three years now. We don't have a dedicated helpdesk team apart for a single part-timer whose responsibility is to organise everyone else and chase people up to make sure the less interesting calls get answered.
Obviously this wouldn't apply everywhere but for our own organisation, we've found it useful to have a rotating policy, where everyone in the tech support team rotates around the desk. Each person averages about half a day to one day per week, including my immediate IT manager. When we have our weekly team meeting, the CIO jumps in and covers the phones for a couple of hours. (He can often help people but if he can't he'll take notes and hand it over.)
The IT team itself is about 15 people, including software developers, security experts, server admins, systems integrators, website admins, project managers, and whoever else we need to get things done. We don't elevate calls in the traditional sense unless an immediate fix is needed and the people on the desk can't handle it, but usually the call will stay open and it won't be long before someone comes along whose direct responsibility corresponds with whatever the problem is.
My primary job is a software developer for mostly in-house software, but I do find it useful to see the software I work on actually getting used, and seeing how badly (or not) it affects people when it breaks in certain ways. I also find it useful having an insight into other people's jobs both on the tech side, as well as actually getting to know the people around the organisation.
The users like it too, because when people call the helpdesk they're usually speaking directly to someone who can actually help them in some way. On average we get on with them really well, and if we want to go out for a group lunch for some reason, it's not hard to find a random person in the organisation who's keen to try covering the phones for a couple of hours.
I'm not sure what practical advantage having a full JVM gives you.
I don't think it'd be much use to me, but perhaps it could be useful if you're trying to port a closed source application or library, maybe? I'm only speculating and I haven't closely looked at either the GWT or at this project.
But... if Microsoft ponies up a few buckazoids and delivers some value to OLPC such that it helps OLPC meets its goals, then, how is that bad for the kids getting the computers, all Windows cracks aside?
A huge advantage of Linux, or anything open source, is that the OLPC developers can tinker with and customise it themselves to match exactly what they want and make it 100% in line with the primary goals of their project. Linux and OSS developers might help, but if they don't it puts no serious restrictions on things because somebody else can do it.
If the OLPC developers can do the same with Windows without interference from Microsoft and if it actually offers a better platform then I agree with you. They can't, however, because Microsoft will definitely place restrictions on what can be done, and Microsoft will definitely exert influence over what the final product looks like, and you can be sure that even if those at the front are genuine, influential and controlling factions within Microsoft will be in it with the primary goal of making sure that children in third world countries get locked into knowing all about Microsoft.
OLPC isn't about laptops for children -- it's about improving education opportunities through providing an educational platform, and everything should be tailored towards that goal. For Microsoft and Windows, anything to do with education is a side-effect, with the primary goal being to make money for and give control of the market to Microsoft.
In defense of Vista (never thought I'd write that!) It's my understanding that UAC is actually aimed at developers.
I agree. A few months ago I had a small and simple program which I really wanted to be able to run in a 'setuid' kind of way, and I discovered that I simply couldn't because under the UAC model Microsoft doesn't allow this. It was annoying at the time since I was already planning to separate the admin program I'd written from everything else. Instead I'd have to run the admin program as a service and set it up that way, which was a bit more effort.
In hindsight though if Microsoft had allowed something equivalent to setuid, it'd simply provide a way for lots of lazy developers and vendors to let their hideously bloated and insecure applications keep working without having to update the code to fit into a better security model.
I still wish that UAC wasn't such a pain for users; there are irritations in there that are simply unnecessary. The worst UAC UI example I've found so far is when trying to copy a file to a privileged location. Before getting an elevation dialog, vista first prompts me with a Yes/No dialog which warns that I'm about to get an elevation dialog, asking me if I want to continue. It's a warning for an upcoming login dialog, and it's one of the craziest UI designs I've come across.
Something more equivalent to sudo would also be nice. It's possible to use 'runas' with an administrator account, but the nice thing about 'sudo' is that it doesn't repeatedly prompt for a password with every single invocation. (At least if this is possible then I haven't figured out how to do it.) I don't know how you can emulate this in a GUI environment with the same security emphasis though, because with 'sudo' the user is at least consciously invoking each command as a superuser when they run it on the command line.
While it's hardly unexpected that Office 2007 document format isn't *cough* ISO compliant, 122k errors for a 60Mb file results into a remarkable ~500 bytes of markup per error.
It depends entirely on context and motivations. I don't like what Microsoft appears to have done here, but if you go out intending to make it look really bad, you could simply pick a small bit of the spec that you happen to know Office 2007 doesn't meet, and then ensure that that particular feature is repeated over and over again in your 60MB document.
As long as they call all non-Chinese media "Western Media" they clearly cannot seriously criticize it in any meaningful way. I mean, Fox News, Slashdot, BBC, FAZ and Corriere della Sera are all part of same group of anti-Chinese conspiracy? With that argument, dear Chinese blogging friends, you are becoming pretty laughable.
If the only media you know is a massive government-controlled conglomerate which echoes the government's opinions by design for the good of the Chinese community, it makes complete sense to assume that media outside China works the same way for other governments.
ODF will win in my eyes when MS Office will read and write it (correctly) without requiring a third party plugin. If it's a standard that 95% of users are unable to use (without conscious effort) then it's of limited usefulness.
That aside, I use ODF for my own things and I'd like it a lot if I could actually give my ODF docs to lots of other people and expect them to be able to open them.
My jaw dropped too to see that South African Microsoft executive claim that.
I'm not convinced he meant it exactly as quoted by Slashdot. In the context of the linked article, he goes on to comment about how "even open-source applications have some form of market model, which incentivises them to continue innovating". With that addition he could easily be referring to the incentives gained by working for the community, getting feedback that makes authors feel good about what they're doing, which to me implies that he does actually realise people do this kind of thing.
I don't understand what he meant by saying people don't do anything for charity, because in my mind it's still charity irrespective of whether the author's getting payback by feeling recognised and important. I also have no idea what his argument has to do with someone having condemned software patents, since he's just implied that open source software seems to be doing perfectly fine without them.
I quite like the idea of the space programme, although I don't pay for the US programme so I don't know if my opinion should be counted much in that respect.
That said, I still haven't seen a great argument as to why it makes more sense to throw massive amounts of money into a space programme when that money could be thrown more directly into developing the useful technologies directly for direct use on Earth, and having less overheads. Just to throw in one example, velcro has been very useful after being developed as part of the space programme many decades ago. But if there had been sufficient funding and an established demand for something that did the same kind of thing, it could have been developed without space.
If anything, I think the space programme has resulted in a lot of new technologies for which the demand outside space travel wasn't immediately obvious. Maybe that's where it's most useful... but it seems a bit far-fetched to use this an an argument to justify spending perhaps hundreds of billions of dollars to visit space when the same money might accomplish the same thing or more without visiting space.
Personally I think space travel is most useful in improving society because it gets people's attention. People will actually support spending hundreds of billions of dollars to visit space, or at least it's easier to convince them. If you try to argue that hundreds of billions (or trillions) of dollars should be spent on non-specific scientific funding to invent cool things, but nothing that anyone's thought of a clearly good reason for yet, then people will start complaining about day-to-day operational problems like overcrowded hospitals.
Is it unusual in your experience for, say, a corporate IT department to destroy hard drives by policy?
Uh, no? It's not unusual at all I don't think, although I'm currently coming from the perspective of a relatively small non-US government department (~350 people).
We have a strict policy that all hard disks from PCs should be destroyed or securely wiped before they leave the building, or they'll remain stored in a locked area and not leave the building until this is possible. This includes photocopiers, printers, scanners and fax machines, and anything with persistent storage. Laptops and tablet PCs which leave the building (some of which go all over the world) are required to have strongly encrypted disks which need a password protected key on an external device to access, and the staff are educated about why that's important and why the key should be stored separately. We don't explicitly use hard disks in PCs for any kind of persistent reliable data storage, but they get used as a matter of course for caching all kinds of things, and this includes whatever people might be working on at the time. (Our document management system caches documents people are working on to the local drive, for good reason.)
Where possible we get broken PCs repaired on site, and when this hasn't been possible in the past we've had issues of the PCs being stolen from the suppliers while they're under maintenance. A lot of them also go to schools once we're done with them, and it makes absolute sense for us to take responsibility and make sure the disks don't get out into the wild with data possibly still on them.
Keep in mind though that this policy is completely separate from the data retention policy that we have. Staff are required to file their documents and emails according to what kind of business it is (we have systems to help them streamline this so there isn't too much overhead), and they get chased up by the document management people (who routinely monitor new documents) if they don't, or if it looks like they're filing things incorrectly. Document accesses and modifications get audited and once a document's been filed, it can't be deleted or changed by anyone except document management staff (or the IT team I guess), and it'd be very difficult if at all possible to do that without leaving an audit trail. It'd also be difficult to delete and modify the backups, some of which are off-site.
Data on the network, intranet and in our document management system gets filed, stored, routinely backed up and sent via a secure courier to a vault in a different city. It might get disposed of after some amount of years, or alternatively sent over to the Archives department if it's that important. Whatever happens gets decided according to whatever retention policy the document management team sets, and that's usually governed by law. As an IT group we don't set the data retention policies for the different files and classes of documents (we're not lawyers or librarians), but we work closely with the people who do.
Having said all of this, I don't think the organisation has any corrupt employees who are habitually trying to manipulate the system and break the rules for their own personal or political gain. If we did, then it might be more difficult to manage, because there's only so much that can be done if someone on the inside isn't trustworthy. But if you're in a well run organisation then those people should be out the door as soon as anything like that is discovered.
In the article, the author describes several uses he had when using Kubuntu. I have had similar issues, but all is reduced to the fact that Kubuntu is a hack "KDE-patched" version of Ubuntu.
Well I was a little disappointed reading Roblimo's review because it looked as if he thought he was reviewing KDE when he was really reviewing Kubuntu's packaged distribution of KDE, which in my experience is quite different. I installed Kubuntu on a new laptop after using KDE on Debian on my desktop PC for a long time, and Kubuntu felt very different. If I'd downloaded and compiled KDE myself, I suspect it would be different from the Debian packaged edition. In particular, I had a difficult time trying to figure out how to configure KDE as I wanted in Kubuntu, because I couldn't easily find many of what I've seen as standard KDE configuration utilities in the past. Kubuntu seemed to want me to use the (K)Ubuntu utilities to configure nearly everything, to the extend where it'd hide (or not even provide) the native configuration managers for the some of the applications.
This is fair enough too, because the distributions themselves are typically what people want to install with something like Ubuntu and Kubuntu. It's easier for the distro to support the desktops if they're all configured in ways provided by the distro. It's also onc of the nice things about open source that there's a huge amount of freedom in the licensing for distro maintainers to change and alter things to work with everything else in their distro, with their distro philosophy, and whatever else they like. This is something you simply don't get with proprietary software where commercial licensing issues tend to trump freedom issues.
Reviewing Kubuntu is not reviewing KDE, just as reviewing KDE isn't the same as reviewing KDE on Kubuntu. People make comparisons between Linux and Windows and KDE and Windows and KDE and Gnome, and Roblimo's just another of these in this case. Kudos to him for at least pointing out that he was using Kubuntu and comparing it with Ubuntu, though. At least there's some context there.
Microsoft just doesn't *understand* open standards
on
India Votes Against OOXML
·
· Score: 5, Informative
The real problem is Microsoft's waffling on making the standard open. If they had unequivocally placed the standard and all necessary patents in the public domain and committed to keeping it stable, more people might vote for it.
Recently I accidentally went to a short promotional Microsoft presentation (non-US) about OOXML for work. From the description about integrating with Office from a programmers' perspective, I'd thought it was going to be about writing Office addins, but it turned out to be a promotional-fest for OOXML in front of about 30 or so local software architechts for various companies and government organisations.
They started with a couple of locals without explaining what was coming -- one guy had built a Silverlight application that could parse basic OOXML Word documents and display them according to the OOXML specification. The other guy had written a web app that generated its own Office 2007 documents (Word and Excel) without having to rely on any third party or binary manipulation.
Then the local Microsoft CIO jumped up, having recently returned from Geneva, and started complaining about how there were really a small segment of people who had gripes with Microsoft and were refusing to work with Microsoft and trying to stop the standard going through for its own sake. They made a big thing about how the two people who'd just presented hadn't needed to read a complete 6000 page specification to do what they'd done, and he used the phrase "defacto standard" in virtually every sentence. They were preaching to the converted on this occasion, considering the room was full of people who were already big Microsoft customers, and really only wanted reassurance rather than to be convinced. I was tempted to ask if Microsoft ever had any plans to support the OASIS standard, but I didn't in the end.
I came away from that presentation with the impression that Microsoft as a company, and especially at the executive level, doesn't actually have a clear understanding of what an Open Standard is. The entire focus of Microsoft is that their Office suite is by far the most popular (for whatever reason), and therefore Microsoft should be the one to decide the standard. If someone else did that while Microsoft was looking the other way, then it must have been an accidental quirk that now needs to be corrected.
Perhaps there's some idea somewhere up in the ranks of leveraging their broken format in the future to reinforce their market dominance should there ever be a problem, but I think for most of them, they're just a bit pissed off or shocked that someone else has already defined a standard and is now trying to tell Microsoft that it can't do what it wants to do. After all, it's not "supposed" to work that way in their minds... Surely the "defacto standard" that's used everywhere should be the one that matters, right?
In their own minds, most of the Microsoft managers are quite certain that Microsoft would never abuse its position, or their already fundamentally of the belief that it's only fair that money should always change hands for these kinds of things, and that if Open Source apps can't find sources of funding then it's their own problem. (Money makes Microsoft go round, after all. It shouldn't be surprising for Microsoft employees to have those kinds of ethics.)
The frustrating addendum to this is that many businesses are in exactly the same mindset as Microsoft because money makes their business go around, too. If Microsoft starts using badly documented parts of their spec and charging for others to implement it, those people will quite happily either keep using Microsoft products, or pay for a product that costs extra as part of the necessity of paying the Microsoft tax. These people haven't even consciously dealt with concepts like standards definitions before, they don't appreciate how critically important it is to get it right, and they don't want to now. That's where Microsoft is getting its support from.
Although they supposedly give more flexibility, something as essential as reading from and writing to a file becomes a hassle with.NET or Java. It's easy to get lost in whether we need a FileInputStream, or whether we should wrap a FileInputReader in a TextInputBuffer, and so forth. Give me fopen() any day.
I've been writing code in a Windows shop using.NET for a couple of years now. I like coding in C at times and still do for some things, but when I'm writing.NET apps I don't really have much of an issue with it. That's what documentation is for, and MSDN's pretty good and up to date. Personally I find the following a lot easier to use than something like fopen():
using(Stream fs = new FileStream("somefile.txt")) {// Do stuff... }
It opens the file, lets you do stuff, and makes absolutely sure it's closed when you're finished irrespective of what happens. When I've tried to write something similar in C it's more direct, but there's a lot more need to think about error conditions rather than what I'm actually trying to do.
I agree with you that sometimes it gets a bit confusing discerning between all the different helper and wrapper classes, but I've usually found there are good reasons for them once I get into using them, and it for me it does make it easier to manage all kinds of potential input without having to care too much. Trying to keep track of all the exceptions that might potentially pop up from within all the wrappers is a real pain, though.
I would of thought the two would of gone hand in hand. How else to the politicians intend to persuade you lot to vote?
I don't know exactly what you guys see inside the US, but near 100% of the media that gets exported outside which I see appears to be much more concerned with who's likely to get elected rather than why people should actually elect them. There's a lot more presentation, smiley faces, crowds of people cheering, and anecdotal notes about things like "black" or "female" or "experience" coming through. There's virtually no information or substance about actual policy. (Keeping in mind that most of what we get is about the Democrat primaries for some reason.)
Perhaps this is just because it's outside the US, even though it'd still be nice to have more information about how various US candidates might effect the rest of the world. (Our local media is hopeless and will happily regurgitate whatever reporting is popular overseas, however. It gets ratings and it's more cost effective than doing their own journalism about things that might be more directly relevant to their own viewers.)
Slashdot Mirror
Short Round was annoying, but nowhere near as irritating as a moronic woman who was screaming and being mocked by the other characters every 10 minutes.
Raiders was an awesome movie and it was one of the few action movies ever to have had a really strong supporting female character who could actually stand up for herself on her own merits, often challenging the main character. When Lucas and the studios tried to capitalise on the first movie's popularity, the Marion character was dropped and much of the really good stuff that made Raiders' was thrown out the window in favour of stupid 1980's movie cliches that were popular at the time. (Notably the annoying kid, and the annoying lady who can't look after herself and needs rescuing by more intelligent, stronger and more competent male characters, including the 10 year old annoying kid at times.)
It looks as if they've re-cast Karen Allen as Marion in the new film (even third on the IMDB list), so I'm hoping there might be something more than a cameo and that it's actually done well. I guess we'll have to wait and see.
Not to mention that this is also great publicity for Seagate.
I'm not personally a proponent of OO-for-everything being practical today, but there are ways that languages can do things nicely if you're willing to think about it in a different mindset. OO is about thinking in terms of objects, not about spinning out annoying extra object instantiations all over the place. (Actually OO is about a lot of abstract and controversial things that go well beyond simply having objects, depending on who you ask.)
If I was trying to design a sin() function in an OO way, I'd probably put a .sin() method on the number class, to return another number object. If this idea is going to require too many thousands of mostly-redundant methods available on every Number object, then perhaps the language (which might allow for custom extensions of existing classes) would require that the programmer import a Math library of some kind before the .sin() method became available on those objects.
I'm not sure they had much choice in all honesty. There are so many VB developers out there that there would have been an outcry if Microsoft hadn't given them some kind of clear path to move to .NET. Furthermore, many of those developers would have avoided moving to .NET at all, and would be writing awful code in just as awful programming frameworks that were no longer supported. At least Microsoft has cleaned up a lot of the worst VB quirks and inconsistencies when they moved it to VB.NET, such that it's now almost possible to write good code in VB.NET if you're so inclined. (Although why would you when C# is so much easier to write with once you understand it?). Bad code tends to be more predictable and do less damage on average.
.NET is nice when you can stay inside .NET, but I've found this is very difficult unless you're developing a relatively simple application. As soon as you want to do anything fancy to interact with the underlying OS or other apps, it usually becomes necessary to start playing with COM wrappers and P/Invoke on at least some level. If you're lucky you might find a third party .NET wrapper library someone's written to do approximately what you want, but quality of those kind of things varies and it usually still requires some interaction with the less-safe APIs.
My experience trying to write a toolbar for use within our organisation would have been horrendous if I hadn't come across BandObjects (or the original), which is clearly popular if you look at the forum below, even though it's not actively maintained by its original author. I still came up with a variety of issues trying to get it to work in Vista, particularly involving getting popup menus that would overlap the Window taskbar properly, and detecting Windows colour themes in a nice way to make sure text was readable. Both of these things need some additional coding that bypasses .NET completely to talk directly to the Windows API using ugly hard-coded constants that mimic definitions in some anonymous Microsoft C++ header file (thus will probably lose compatibility sooner or later), and I still haven't gotten them to work nicely.
There does seem to be a certain irony in writing to Mr Zelnick's mother to complain about his upbringing when Jack Thompson's main argument, as far as I can gather, is that violent media, including video games, are primarily responsible for violence in young people and society.
If that's the case, why does he expect that she had anything to do with it? Surely she's as much a victim of a free state as all the other parents whose children are running wild and uncontrolled.
I can think of at least two office suites (possibly three) off the top of my head which either use ODF as their primary document formats, or plan to in the near future. They've all done so because it makes sense for users to be saving in an interchangable and properly standardised format, and it means that somebody's application suite doesn't have to be decided by their archive of documents they have to be able to reliably open. Unfortunately the number of implementing office suites doesn't equate to widespread usage because the provider of the most popular Office suite, which controls the market and locks people in, has strong priorities that completely ignore what's best for customers.
Microsoft will never natively support ODF if they have anything to say about it, even if it is best for their customers. This is because their own unimplementable formats give it not just control over individual customers, but a virally* distributive control over everyone with whom their customers exchange documents. Consequently the most popular Office suite will always be decided by Microsoft's lock-in tactics rather than which suite is actually best for customers.
I'm sorry but if Microsoft's skewed priorities have led it to interfere with government processes and corrupt them at everyone else's expense, simply so it can keep control of the market and continue its monopoly, then I'm fully in favour of complaining and making it known as clearly and noisily as possible and every level where it occurred.
* If Microsoft can refer to the GPL as viral then I'm going to do the same for Office.
It's a shame he was billed for it and perhaps it wasn't unfair and should be considered that way, but any head injury is serious until proved otherwise, regardless of what the person who suffered it might think. It sounds as if the ski patrol people on-site were doing their job properly as far as looking after him. If you get knocked unconscious then you potentially have concussion, internal bleeding or swelling which can have non-immediate effects, and you need to get to a hospital as soon as possible to be checked out properly.
If I was out in a wilderness area and responsible for someone who was knocked unconscious, I'd definitely be getting them evacuated as soon as it was possible to make arrangements, irrespective of how quickly they woke up afterwards or what they said. If calling in a helicopter was the fastest way out, they'd get it whether they wanted it or not. It's basic first aid. It's true that sometimes this gets overlooked, especially for drunk people on the streets (to name an example), but it's especially important in any remote area where you could be a long way from help if you suddenly drop to the ground from an injury you didn't know you still had, because if that happens then you're likely to die.
I'd presume that it's more of a claim on Steve Fossett's fortune rather than directly billing his widow, although I'm not a lawyer and I'm also not informed about whatever arrangements Steve Fossett had with the state of Nevada before he left.
As for taxpayer funding, I'm in mixed minds about whether people should be charged for their rescues. In New Zealand, and I'm sure many other places, we get lots of tourists who come here for the outdoor experience. Most of them are great, but there are still a lot of tourists who completely ignore all the advice about what outdoor safety gear they need and where they should and shouldn't go according to their experience, and consequently need to be rescued out of caves or off the side of mountains or from behind flooded rivers or wherever else they've gotten themselves stuck.
Rescue helicopters and their crews are expensive, and large volunteer search parties typically mean that people are giving up their daytime salaries to go and look for someone who got themselves into trouble. I can appreciate that sometimes things just go wrong, but if a person put themselves at risk through their own fault and triggered a rescue operation, I don't personally have as much of an issue in sending them the bill, or at least charging them an impressionable portion of the bill if it's unrealistic. (Sending a bill that will ruin someone for life isn't much help at all.)
Steve Fossett has a history of putting himself at obvious risk as part of his hobby. This is fine if it's what he wants to do, but when he triggers a massive rescue operation I can appreciate that some people might think he should be responsible for footing at least part of the bill, if not the whole thing. Whether that bill should be transferred to the person who's inheriting his vast fortune would, I think, depend on legal details.
I'd have thought it'd look more like:
I agree. I've been working in a medium sized organisation (~400 people, all in one building) for about three years now. We don't have a dedicated helpdesk team apart for a single part-timer whose responsibility is to organise everyone else and chase people up to make sure the less interesting calls get answered.
Obviously this wouldn't apply everywhere but for our own organisation, we've found it useful to have a rotating policy, where everyone in the tech support team rotates around the desk. Each person averages about half a day to one day per week, including my immediate IT manager. When we have our weekly team meeting, the CIO jumps in and covers the phones for a couple of hours. (He can often help people but if he can't he'll take notes and hand it over.)
The IT team itself is about 15 people, including software developers, security experts, server admins, systems integrators, website admins, project managers, and whoever else we need to get things done. We don't elevate calls in the traditional sense unless an immediate fix is needed and the people on the desk can't handle it, but usually the call will stay open and it won't be long before someone comes along whose direct responsibility corresponds with whatever the problem is.
My primary job is a software developer for mostly in-house software, but I do find it useful to see the software I work on actually getting used, and seeing how badly (or not) it affects people when it breaks in certain ways. I also find it useful having an insight into other people's jobs both on the tech side, as well as actually getting to know the people around the organisation.
The users like it too, because when people call the helpdesk they're usually speaking directly to someone who can actually help them in some way. On average we get on with them really well, and if we want to go out for a group lunch for some reason, it's not hard to find a random person in the organisation who's keen to try covering the phones for a couple of hours.
I don't think it'd be much use to me, but perhaps it could be useful if you're trying to port a closed source application or library, maybe? I'm only speculating and I haven't closely looked at either the GWT or at this project.
A huge advantage of Linux, or anything open source, is that the OLPC developers can tinker with and customise it themselves to match exactly what they want and make it 100% in line with the primary goals of their project. Linux and OSS developers might help, but if they don't it puts no serious restrictions on things because somebody else can do it.
If the OLPC developers can do the same with Windows without interference from Microsoft and if it actually offers a better platform then I agree with you. They can't, however, because Microsoft will definitely place restrictions on what can be done, and Microsoft will definitely exert influence over what the final product looks like, and you can be sure that even if those at the front are genuine, influential and controlling factions within Microsoft will be in it with the primary goal of making sure that children in third world countries get locked into knowing all about Microsoft.
OLPC isn't about laptops for children -- it's about improving education opportunities through providing an educational platform, and everything should be tailored towards that goal. For Microsoft and Windows, anything to do with education is a side-effect, with the primary goal being to make money for and give control of the market to Microsoft.
I agree. A few months ago I had a small and simple program which I really wanted to be able to run in a 'setuid' kind of way, and I discovered that I simply couldn't because under the UAC model Microsoft doesn't allow this. It was annoying at the time since I was already planning to separate the admin program I'd written from everything else. Instead I'd have to run the admin program as a service and set it up that way, which was a bit more effort.
In hindsight though if Microsoft had allowed something equivalent to setuid, it'd simply provide a way for lots of lazy developers and vendors to let their hideously bloated and insecure applications keep working without having to update the code to fit into a better security model.
I still wish that UAC wasn't such a pain for users; there are irritations in there that are simply unnecessary. The worst UAC UI example I've found so far is when trying to copy a file to a privileged location. Before getting an elevation dialog, vista first prompts me with a Yes/No dialog which warns that I'm about to get an elevation dialog, asking me if I want to continue. It's a warning for an upcoming login dialog, and it's one of the craziest UI designs I've come across.
Something more equivalent to sudo would also be nice. It's possible to use 'runas' with an administrator account, but the nice thing about 'sudo' is that it doesn't repeatedly prompt for a password with every single invocation. (At least if this is possible then I haven't figured out how to do it.) I don't know how you can emulate this in a GUI environment with the same security emphasis though, because with 'sudo' the user is at least consciously invoking each command as a superuser when they run it on the command line.
It depends entirely on context and motivations. I don't like what Microsoft appears to have done here, but if you go out intending to make it look really bad, you could simply pick a small bit of the spec that you happen to know Office 2007 doesn't meet, and then ensure that that particular feature is repeated over and over again in your 60MB document.
If the only media you know is a massive government-controlled conglomerate which echoes the government's opinions by design for the good of the Chinese community, it makes complete sense to assume that media outside China works the same way for other governments.
As someone who lives outside the USA, I'm at least as concerned about how to enforce USA-based companies from tracking and annoying me.
ODF will win in my eyes when MS Office will read and write it (correctly) without requiring a third party plugin. If it's a standard that 95% of users are unable to use (without conscious effort) then it's of limited usefulness.
That aside, I use ODF for my own things and I'd like it a lot if I could actually give my ODF docs to lots of other people and expect them to be able to open them.
I'm not convinced he meant it exactly as quoted by Slashdot. In the context of the linked article, he goes on to comment about how "even open-source applications have some form of market model, which incentivises them to continue innovating". With that addition he could easily be referring to the incentives gained by working for the community, getting feedback that makes authors feel good about what they're doing, which to me implies that he does actually realise people do this kind of thing.
I don't understand what he meant by saying people don't do anything for charity, because in my mind it's still charity irrespective of whether the author's getting payback by feeling recognised and important. I also have no idea what his argument has to do with someone having condemned software patents, since he's just implied that open source software seems to be doing perfectly fine without them.
I quite like the idea of the space programme, although I don't pay for the US programme so I don't know if my opinion should be counted much in that respect.
That said, I still haven't seen a great argument as to why it makes more sense to throw massive amounts of money into a space programme when that money could be thrown more directly into developing the useful technologies directly for direct use on Earth, and having less overheads. Just to throw in one example, velcro has been very useful after being developed as part of the space programme many decades ago. But if there had been sufficient funding and an established demand for something that did the same kind of thing, it could have been developed without space.
If anything, I think the space programme has resulted in a lot of new technologies for which the demand outside space travel wasn't immediately obvious. Maybe that's where it's most useful... but it seems a bit far-fetched to use this an an argument to justify spending perhaps hundreds of billions of dollars to visit space when the same money might accomplish the same thing or more without visiting space.
Personally I think space travel is most useful in improving society because it gets people's attention. People will actually support spending hundreds of billions of dollars to visit space, or at least it's easier to convince them. If you try to argue that hundreds of billions (or trillions) of dollars should be spent on non-specific scientific funding to invent cool things, but nothing that anyone's thought of a clearly good reason for yet, then people will start complaining about day-to-day operational problems like overcrowded hospitals.
Uh, no? It's not unusual at all I don't think, although I'm currently coming from the perspective of a relatively small non-US government department (~350 people).
We have a strict policy that all hard disks from PCs should be destroyed or securely wiped before they leave the building, or they'll remain stored in a locked area and not leave the building until this is possible. This includes photocopiers, printers, scanners and fax machines, and anything with persistent storage. Laptops and tablet PCs which leave the building (some of which go all over the world) are required to have strongly encrypted disks which need a password protected key on an external device to access, and the staff are educated about why that's important and why the key should be stored separately. We don't explicitly use hard disks in PCs for any kind of persistent reliable data storage, but they get used as a matter of course for caching all kinds of things, and this includes whatever people might be working on at the time. (Our document management system caches documents people are working on to the local drive, for good reason.)
Where possible we get broken PCs repaired on site, and when this hasn't been possible in the past we've had issues of the PCs being stolen from the suppliers while they're under maintenance. A lot of them also go to schools once we're done with them, and it makes absolute sense for us to take responsibility and make sure the disks don't get out into the wild with data possibly still on them.
Keep in mind though that this policy is completely separate from the data retention policy that we have. Staff are required to file their documents and emails according to what kind of business it is (we have systems to help them streamline this so there isn't too much overhead), and they get chased up by the document management people (who routinely monitor new documents) if they don't, or if it looks like they're filing things incorrectly. Document accesses and modifications get audited and once a document's been filed, it can't be deleted or changed by anyone except document management staff (or the IT team I guess), and it'd be very difficult if at all possible to do that without leaving an audit trail. It'd also be difficult to delete and modify the backups, some of which are off-site.
Data on the network, intranet and in our document management system gets filed, stored, routinely backed up and sent via a secure courier to a vault in a different city. It might get disposed of after some amount of years, or alternatively sent over to the Archives department if it's that important. Whatever happens gets decided according to whatever retention policy the document management team sets, and that's usually governed by law. As an IT group we don't set the data retention policies for the different files and classes of documents (we're not lawyers or librarians), but we work closely with the people who do.
Having said all of this, I don't think the organisation has any corrupt employees who are habitually trying to manipulate the system and break the rules for their own personal or political gain. If we did, then it might be more difficult to manage, because there's only so much that can be done if someone on the inside isn't trustworthy. But if you're in a well run organisation then those people should be out the door as soon as anything like that is discovered.
Well I was a little disappointed reading Roblimo's review because it looked as if he thought he was reviewing KDE when he was really reviewing Kubuntu's packaged distribution of KDE, which in my experience is quite different. I installed Kubuntu on a new laptop after using KDE on Debian on my desktop PC for a long time, and Kubuntu felt very different. If I'd downloaded and compiled KDE myself, I suspect it would be different from the Debian packaged edition. In particular, I had a difficult time trying to figure out how to configure KDE as I wanted in Kubuntu, because I couldn't easily find many of what I've seen as standard KDE configuration utilities in the past. Kubuntu seemed to want me to use the (K)Ubuntu utilities to configure nearly everything, to the extend where it'd hide (or not even provide) the native configuration managers for the some of the applications.
This is fair enough too, because the distributions themselves are typically what people want to install with something like Ubuntu and Kubuntu. It's easier for the distro to support the desktops if they're all configured in ways provided by the distro. It's also onc of the nice things about open source that there's a huge amount of freedom in the licensing for distro maintainers to change and alter things to work with everything else in their distro, with their distro philosophy, and whatever else they like. This is something you simply don't get with proprietary software where commercial licensing issues tend to trump freedom issues.
Reviewing Kubuntu is not reviewing KDE, just as reviewing KDE isn't the same as reviewing KDE on Kubuntu. People make comparisons between Linux and Windows and KDE and Windows and KDE and Gnome, and Roblimo's just another of these in this case. Kudos to him for at least pointing out that he was using Kubuntu and comparing it with Ubuntu, though. At least there's some context there.
Recently I accidentally went to a short promotional Microsoft presentation (non-US) about OOXML for work. From the description about integrating with Office from a programmers' perspective, I'd thought it was going to be about writing Office addins, but it turned out to be a promotional-fest for OOXML in front of about 30 or so local software architechts for various companies and government organisations.
They started with a couple of locals without explaining what was coming -- one guy had built a Silverlight application that could parse basic OOXML Word documents and display them according to the OOXML specification. The other guy had written a web app that generated its own Office 2007 documents (Word and Excel) without having to rely on any third party or binary manipulation.
Then the local Microsoft CIO jumped up, having recently returned from Geneva, and started complaining about how there were really a small segment of people who had gripes with Microsoft and were refusing to work with Microsoft and trying to stop the standard going through for its own sake. They made a big thing about how the two people who'd just presented hadn't needed to read a complete 6000 page specification to do what they'd done, and he used the phrase "defacto standard" in virtually every sentence. They were preaching to the converted on this occasion, considering the room was full of people who were already big Microsoft customers, and really only wanted reassurance rather than to be convinced. I was tempted to ask if Microsoft ever had any plans to support the OASIS standard, but I didn't in the end.
I came away from that presentation with the impression that Microsoft as a company, and especially at the executive level, doesn't actually have a clear understanding of what an Open Standard is. The entire focus of Microsoft is that their Office suite is by far the most popular (for whatever reason), and therefore Microsoft should be the one to decide the standard. If someone else did that while Microsoft was looking the other way, then it must have been an accidental quirk that now needs to be corrected.
Perhaps there's some idea somewhere up in the ranks of leveraging their broken format in the future to reinforce their market dominance should there ever be a problem, but I think for most of them, they're just a bit pissed off or shocked that someone else has already defined a standard and is now trying to tell Microsoft that it can't do what it wants to do. After all, it's not "supposed" to work that way in their minds... Surely the "defacto standard" that's used everywhere should be the one that matters, right?
In their own minds, most of the Microsoft managers are quite certain that Microsoft would never abuse its position, or their already fundamentally of the belief that it's only fair that money should always change hands for these kinds of things, and that if Open Source apps can't find sources of funding then it's their own problem. (Money makes Microsoft go round, after all. It shouldn't be surprising for Microsoft employees to have those kinds of ethics.)
The frustrating addendum to this is that many businesses are in exactly the same mindset as Microsoft because money makes their business go around, too. If Microsoft starts using badly documented parts of their spec and charging for others to implement it, those people will quite happily either keep using Microsoft products, or pay for a product that costs extra as part of the necessity of paying the Microsoft tax. These people haven't even consciously dealt with concepts like standards definitions before, they don't appreciate how critically important it is to get it right, and they don't want to now. That's where Microsoft is getting its support from.
I've been writing code in a Windows shop using .NET for a couple of years now. I like coding in C at times and still do for some things, but when I'm writing .NET apps I don't really have much of an issue with it. That's what documentation is for, and MSDN's pretty good and up to date. Personally I find the following a lot easier to use than something like fopen():
using(Stream fs = new FileStream("somefile.txt")) { // Do stuff... }
It opens the file, lets you do stuff, and makes absolutely sure it's closed when you're finished irrespective of what happens. When I've tried to write something similar in C it's more direct, but there's a lot more need to think about error conditions rather than what I'm actually trying to do.
I agree with you that sometimes it gets a bit confusing discerning between all the different helper and wrapper classes, but I've usually found there are good reasons for them once I get into using them, and it for me it does make it easier to manage all kinds of potential input without having to care too much. Trying to keep track of all the exceptions that might potentially pop up from within all the wrappers is a real pain, though.
I don't know exactly what you guys see inside the US, but near 100% of the media that gets exported outside which I see appears to be much more concerned with who's likely to get elected rather than why people should actually elect them. There's a lot more presentation, smiley faces, crowds of people cheering, and anecdotal notes about things like "black" or "female" or "experience" coming through. There's virtually no information or substance about actual policy. (Keeping in mind that most of what we get is about the Democrat primaries for some reason.)
Perhaps this is just because it's outside the US, even though it'd still be nice to have more information about how various US candidates might effect the rest of the world. (Our local media is hopeless and will happily regurgitate whatever reporting is popular overseas, however. It gets ratings and it's more cost effective than doing their own journalism about things that might be more directly relevant to their own viewers.)