Since when did memory usage become such a big deal?
Memory is a big deal for me because I have a cheap laptop with as much RAM as it can hold. Still RAM can be the limiting factor for two reasons: First, I often have another OS running in a VM, which really puts a dent in your available RAM. Second, My OS and most user software have stabilized and suspend modes have become fast enough. Things just don't crash anymore and as a result I just don't reboot anymore. My uptime on my laptop is generally the same as a few days after the last security patch was released. Sometimes this is months.
Really I don't care if Firefox uses twice as much memory as it does during the first hour I use it. The problem comes when I open a few tabs of reference material, then leave those tabs open in the background until I get back to them. In this instance I've gone back to Firefox to kill it, because it gradually used more and more RAM until other applications were starved. The only other programs I've used that exhibited that behavior are Adobe InDesign (which is even worse than Firefox, 2 days between restarting it, tops) and MS Word (which is about the same as Firefox.
I mean Firefox has had some nasty memory leaks for the longest time and absolutely I would love to see those fixed. But it seems like this is more than just that, it seems like some big epeen contest between browsers.
I know it is uncommon in the computing industry these days, but it is called 'competition'. There are several viable alternatives for your browser of choice and each is striving to outdo the others and gain market share. Sadly, this competition does not extend to Microsoft who is still abusing their monopoly to take most of the market. The end result, however, is that among other browsers a lot of the market is savvy enough to see what memory usage is easily and hence use it as a benchmark of sorts.
...or one like IE 7 which is more polished than both Firefox and Opera.
Heh, your milage many vary, caveat emptor, this statement does not reflect the views of this station or geekdom, etc.
Ugh. Please, just admit you hate MS, and you want them to fail at everything...
But I don't hate MS. I do hate what they've done to the science of computing. I don't want them to fail as a company, but I do want them to fail at their illegal business practices.
...and you're[sic] zealotry wants ODF to succeed...
I want ODF to succeed and I want the current version of OOXML to fail, because I want the office suite market to be competitive and to actually advance. I want Microsoft to have to make MS Office the best offering on the market and keep improving it to keep it the best if they want to have a large market share. I have no problem with MS Office being the most widely used office suite, but I want MS to have to earn that status by making it better than everything else, not by breaking compatibility with others and making it hard for users to use anything else because of their shenanigans.
...and that you're worried that now the OOXML is a standard
No, now I'm worried that purchasing agents will think it is a standard and will accept it as if it was a standard, not realizing it won't bring them the advantages a real standard like ODF can.
...the huge install base will put a lot more implementions of OOXML out there than ODF.
Yes that is exactly right. MS's huge market share will put a lot of implementations out there compared to ODF. The fact that you don't recognize leveraging monopoly influence as a bad thing is very sad. I have no problem with OOXML being implemented more than ODF, but it should be because it is a better standard, not because Microsoft is pushing it on people who generally don't have the power to do anything about it.
Kompose is only one of the expose-type features on linux. Another is the expose-like feature in Compiz, enabled by default in Ubuntu. Kubuntu is a community-supported distro & doesn't develop as quickly as Ubuntu.
Really, you can activate an expose like feature in a default Ubuntu install. Do tell me how, since I recently switched to Ubuntu as my main Linux desktop. What mouse button do you press? What key combo do you press? I saw no mention of this anywhere in the help system and have not yet got around to searching for a 3rd party solution.
So - according to you, 'linux' developers aren't interested in zeroconf, but you even point out in your own post that the XO linux developers do support ubiquitous zeroconf.
XO is not a mainstream Linux distro. None of the major Linux distros has made ubiquitous use of zeroconf. Sure, a few developers on a special project aimed at kids did. Everyone else has largely ignored it. Why isn't it used in Ubuntu and Fedora and Suse? Because most Linux on the desktop developers did not bother to copy OS X in this regard. That is a serious failing. Unless you're suggesting that the XO laptop constitutes a viable choice for me to use as my desktop, my point stands.
I misread that as SystemStartup - my apologies, linux indeed does not have something like system services - this is the consequence of being a bazaar rather than cathedral project.
As I said, there is no one person who can push major changes, so in general they don't happen. This is both a strength and weakness for Linux. Pointing out where it is a weakness should not require me to go through several revisions of comments with people who claim to know enough to compare said strengths and weaknesses.
Oooookay? Again - you said "What I find interesting is that major Linux developers don't copy those features from OS X." - not "What I find interesting is that major Linux developers don't copy those features from OS X - and instantly have working implementations"
The point is, while someone is "working on it" for many major distributions, we have no idea if those projects will ever be integrated into the mainstream distributions. There are 10 projects on sourceforge for every one that becomes usable. It has been 3 years since Apple shipped a working version of LaunchD and made the code available. None of the major Linux distros has managed to copy or create their own and ship it in their stable release. At this point, I have my doubts that any will for a long time.
So they're trying to implement it, but thats still not good enough for you.
That is correct. I want there to be enough linux developers familiar with OS X features such that they realize the advantages of those features and prioritize them. Sadly, I see just the opposite. I know a lot of Linux developers, but a good portion of them now use OS X as their desktop and actually oppose cloning OS X features in Linux, not because they aren't useful (they use them daily) but because they aren't useful for a server, which is what they want to use Linux for.
Recent OS X switchers tend to come from Windows via Linux (for a month or two) & don't really have 'real experience' with Linux.
Yeah, that is not at all my experience. I've spent the last 8 years working in various companies that make heavy use of Linux and *BSD in making servers and routers. 8 years ago I knew one or two Mac users. Last year I left a small company of a few hundred developers and more than half were using OS X as their desktop to develop Linux and software that ran on top of Linux. All of them had significant experience with Linux or we would never have hired them.
Furthermore, if we're swapping anecdotes, the closest I've come to people with real experience using both are people who actually use both daily. I'm one of them (along with windows).
Sigh. Why is explaining deficiencies in Linux like pulling teeth? Seriously, why are Linux users in general so opposed to understanding ways someone else has done better and trying to improve based upon them? I'll go through your bullet points one by one.
Zeroconf - Avahi
Yes, there are certainly zeroconf implementations on Linux. My point was "ubiquitous zeroconf." You now what 'ubiquitous' means, right? Both Pidgin and Kopete support zeroconf for discovering local chat clients. Aside from that and a general zeroconf browser in Ubuntu, they don't make use of zeroconf for much of anything else. They don't use it to discover terminal servers or remote GUI sessions or shared streaming music. Niether Ubuntu or Kubuntu finds my printer, despite it advertising itself via zeroconf. About the only Linux distro I know of that makes what I'd refer to as ubiquitous use of zeroconf is the XO laptop. Seriously, all major Linux distros I've used have ignored the potential uses. I don't even know of a multi-user text editor that uses zeroconf on Linux. For servers, I can see firewalling it off by default, but for the average desktop, on the local network, it makes sense for the huge functionality win.
Expose - Kompose
Yes, somebody cloned Expose for KDE. It even works fairly well (at least on my Kubuntu box). I still had to go out and search for it and then add the package by hand. That simply isn't going to happen for most Linux newbies, or even Linux users who haven't seen it on OS X. It is nice that I can add such a thing, but this does nothing to convince me the Linux on the desktop developers are willing to adopt real innovations from OS X. Ask me again when it is enabled by default on a major distro.
System services - errr, that's been replaced by Launchd on your list.
You see. This is exactly what I'm talking about. You don't even know what "system services" are on OS X. How can I believe you are making an informed comparison, when you obviously have never taken the time to learn what OS X's real strengths are. I've had this same conversation with a baker's dozen Linux on the desktop developers. None of them knew what this feature was and once explained they all agreed it would be useful... but too much work to try by themselves.
System services on OS X allow functionality to be offered by plug-ins and applications to other applications, universally, without the developer of the other application doing any additional work. It is how OS X offers system wide functions like spell checking and grammar checking, but also allows the user to add arbitrary functionality on par with those. For example, I long ago added a service to translate between german and english, and now it is a key combo away in my mail client, chat client, and web browser. When I installed Graphviz it offered its service to other applications so in any other program (terminal, textedit, excel) I can just highlight data and hit a key combo and it sicks it into Graphviz and creates a graph of the data, which it loads into my copy/paste buffer. I don't have to bother wasting time opening the program and copying and pasting data and rendering it. I probably use a dozen different system services in my average work day for spelling, grammar, bibliography auto-formatting, dictionary/thesaurus/wikipedia lookup, text manipulation scripts, etc. When I did development work more, I used a different set of them. The whole point is it allows the user to customize the functionality in programs, without having to write any code and recompile, or even have the source to a given application. No Linux distro I've tried comes close and no KParts is not "just as good" or even the same concept.
Launchd - ReplacementInit - Considered Apple's LaunchD and Sun's SMF amongst others, but none were quite the right fit.
Yes, every distro has such a project and as far as I know none of them are compatible (except as backward compatible mode
Which "they" to you mean here? ISO (and to a lesser extent, Ecma) would be the proper "they" in this case - not Microsoft - and you never meant to imply otherwise, I hope.
Actually I did mean to imply MS should have provided a version of the spec that solves all the major problems noted with it by others.
Incorporating all of the BRM recommended changes into a usable standard is something that's being worked on right now, and will continue if DIS2900 has in fact passed this vote.
That's an interesting theory. I take it you've never heard of Microsoft before today?
It has already been agreed that the parts which reference Word97 (which were always clearly marked as deprecated and optional) will be moved to a new section - also clearly marked 'deprecated and optional'.
That's not good enough. If it is included in the spec even as an optional component, MS can implement it and their documents can break when other programs that also conform to the spec (but don't implement optional portions) open them. This undermines the point of having a standard.
I guess you're going by the SFLC's recent remarks about the OSP here. Oddly they didn't feel the same way about remarkably similar covenants from IBM and Sun.
That's not odd at all. With Sun and IBM it is academic because neither has monopoly influence on the market and as such can't undermine competition by those means. MS does have monopoly influence in the office suite market (although they've spent a buttload of money settling lawsuits trying to make sure no court makes that declaration). When a document breaks when written by IBM's software, people can avoid using IBM's software or users can complain and IBM will fix it. When the same thing happens with documents saved by MS Office, because of its overwhelming market share, it doesn't matter who caused the document to break, it will always be treated by the market as a flaw in the program that is not from MS, thus introducing artificial problems in competitors and undermining the operation of the capitalist free market.
There would seem to be some dissent within SFLC as well, given that Mark Webbink (of SFLC's board, and also speaking as general counsel for Red Hat) has said that he feels the OSP does indeed allow the flexibility to be implemented under F/OSS licenses.
Of course there is dissent, that is why there is a voting process. It is just that the voting process was not designed to be properly robust in the face of a big company with a lot of money and influence trying to undermine the process.
If you've been involved with the writing of some ISO standards, and you're an advocate of free and open standards, surely it rankles you that many ISO standards aren't freely available, but instead must be purchased for hundreds of dollars ($350 or so for ISO/IEC 26300, a/k/a ODF)?
Not really. Open source or even Free (as in freedom) is not the same thing as free (as in beer). I'm an advocate of open standards because they allow free markets to operate efficiently. That is efficiently sans intervention by an abusive monopolist which has enough political influence and legal savvy to render the courts unable to effectively stop its abuse.
But I digress (sorry) - I have personally looked at both ODF and OOXML. Haven't just taken the pundits opinions on faith - I went and looked. Both have large sections which require much re-reading if one wants to create a good implementation.
Did you perhaps notice that every function in OOXML simply describes how MS Office or the largest third party vendor that interoperates does things. Did you perhaps, notice that the format eschews consistency within itself for the sake of more closely describing what the various component programs of MS Office does? ODF is not perfect, but it sure doesn't take a genius to see the vastly different mindsets appli
OSS is widely used on the desktop, probably more widely than Macintosh. You're confusing the desktop with the home market.
While numbers are hard to calculate for OSS desktop OS's, I haven't seen numbers to support your assertion. That isn't too important though because you are misunderstanding the market distinctions. Apple doesn't sell desktop OS's. They use vertical bundling to bypass the desktop OS market and instead compete in the computer system market. OSS OS's have a very small market share pre-installed or even OS-less whiteboxes.
What does that have to do with anything?
It speaks to the viability of NeoOffice in professional use as a competitor to Apple's office suite. I've seen it there but not on a large scale, not even on the scale of iWork (let alone MS Office).
Well, hey, no need to buy iWork then, right? Don't make me laugh.
Umm, I don't see what you're trying to get at. You argued that Apple as a company was opposed to letting ODF gain a foothold. When I countered with Apple's support in TextEdit you claimed that maybe they only wanted users to be able to read the occasional ODF document, but at the same time keep it from being viable as a competitor. Apple's support for writing ODF contradicts that theory. Claiming now that it isn't important on the third iteration of this subject is weak. Just concede the point gracefully and move on.
Yes, and they could have adopted OpenOffice XML at the time
No, they couldn't have since it wasn't finished. They could have moved to a draft of it and tried to influence the spec enough to work for iWork and OpenOffice, but there is no guarantee they would have been successful. Of course at that point the developers may not have even heard of it let alone considered it viable. We don't even know if they started from scratch with iWork or used an existing code base from Apple or Next. In short, Apple could have considered ODF and possibly adopted it, but we have no idea if they evaluated it or if they did, how risky they considered such a move or how much more expensive it would have been for them. Remember, at that time the spec was not implemented by anyone and there was no guarantee it would be. Further, the spec was at that point completely under Sun's control. For all they knew at that point it was never going to be implemented by anyone else.
Instead, they went through two poorly designed XML formats that were homegrown.
That's an interesting assertion. How, exactly, do you find Apple's XML formats to be poorly designed (and for that matter which formats are you referring to, iWork has 3 formats:.key,.pages, and.numbers).
Apple has a long history of intentional incompatibilities
That's very interesting, but what does that have to do with whether or not they are intentionally making iWork incompatible? At the time of Apple's creation of iWork, there was no approved open standard for office suites and not even OpenOffice would support ODF (when it was finished) for another year.
Apple has no interest in helping ODF adoption or risking their relationship with Microsoft.
Great two more assertions. You do know you have to do more than make assertions in order to make a persuasive argument, right? How about backing up some of your assertions with some facts or logic. Everything you've presented so far has been an empty assertion or a fact I have demonstrated is false or does not prove what you hoped it did. In logic, facts precede forming an opinion. Why don't schools teach logic anymore?
Apple made a deal with Microsoft and they have Microsoft Office on their platform. That's something that helps them a lot in competing with open source desktop operating systems. Why would they want to lose that advantage by supporting ODF?
Apple doesn't really compete with OSS in the desktop space. It might in the future, but to date OSS does not have enough market share to count. Apple does compete with Microsoft and with Microsoft's customers. Microsoft has been leveraging their office suite to gain an advantage in other markets. Ceding the office suite market to MS hurts Apple's iWork sales. It hurts their computer sales too. MS having monopoly influence in the office suite market, but not leveraging it directly against Apple in the computer market is where Apple is now. They know they would be in a better position yet if they could get rid of MS's monopoly influence in the office suite market, especially since MS has announced changes that will lead to their exploiting it against apple (dropping vb scripts in office for the mac).
They can do that with NeoOffice.
Currently, I wouldn't recommend it. NeoOffice is just too slow and buggy for large scale production use (IMHO). In any case, that only applies to sales of computer systems, not to sales of office suites. Apple sells an office suite called iWork, which currently is not in the running for those contracts.
Makes perfect sense: they want people to be able to read the occasional ODF file, but they don't want people to strengthen ODF by fully supporting it.
Except TextEdit has both read and write support for ODF. If they wanted to keep ODF from competing with them, why would they provide support for writing to it? Also, TextEdit ships with every OS X system, iWork does not. That means OS X itself has better support for ODF, by default, than Apple's own iWork formats.
Well, you can "disagree" all you want, but it isn't born out by the facts.
I think it is.
Apple has spent a lot of time and money developing two different XML formats for iWork when they could have just adopted the ODF format.
The first reports of iWork in development were in 2003. It was officially announced in 2004 and was actually available in 2005. ODF 1.0 was finalized in 2006. Draft versions of ODF existed as early as 2002. For reference OpenOffice.org did not support ODF until 2005, the same year iWork was already on sale. It did not gain momentum among any other vendors (except Sun) for another year.
If the iWork team at Apple were aware of the upcoming ODF standard, they could have joined the group and helped create the first version (and used it in Pages). As it was though, Pages was not being targeted as a Word replacement for general word processing. Rather it was targeted as a layout application for home users, sort of like InDesign for home users. Many of the features it used were targeted at that market segment and were quite different and even if Apple had joined there is no guarantee they would have had enough influence to make ODF appropriate for their target market. Even today the committee is pretty much Sun and IBM employees. There are a few Novell people and Chinese government people, two guys from Adobe, and one guy from Google. That's pretty much it. It would be quite nice if Apple joined and provided support in their applications. They may, eventually, do that, especially if the formats see wider uptake in the future. Claiming that Apple is intentionally avoiding compatibility is a stretch. It is probably mostly a matter of which types of employees have influence on the iWork project. It is probably old school Apple guys (who tend to largely ignore standards and security) as opposed to former Next employees and people hired in from the Linux/BSD/Unix community.
Why? Are you saying no one could possibly implement it?
No one, not even MS can implement it because they haven't published a version that makes the required changes, it is ambiguous, it references other formats that are not published, and it contradicts itself. Basically, it is just like the existing MS Office formats. You can't follow a standard. You just have to try to reverse engineer what MS did for the common cases in the most popular version of MSOffice and ignore the uncommon cases.
So what? I don't recall ISO standards as having to be free of patents.
In the Office suite industry two of the four biggest competitors are OSS projects. Creating standards that for legal reasons are not implementable by such a huge portion of the market, undermines the standards process. This is MS's normal tactic, copy a feature of other software, but at the same time undermine the benefits that feature brings to users. It is creating a hollow bullet point for salespeople while not benefiting users in a way that might allow them to choose anything other than MS.
Have you read other ISO standards?
I have. I've even been involved in writing some of them. I dare you to go look at ODF and then look at OOXML. The difference is night and day, even for a layperson.
If it were approved, it would be a standard..
No it wouldn't. Some committee rubber stamping something does not make it a standard. ISO's purpose is not to make an existing format a standard, but to help create and certify that formats are standards. By them certifying OOXML as a standard does not make it one; it just fools people who have relied upon ISO's reputation into thinking it is one. OOXML will not be any more usable or implementable by others or clearer next week after ISO announces they have approved it than it is today. It is just false advertising bought by Microsoft. Strangely some people here object to false advertising, especially when MS has destroyed the credibility and integrity of a standards body to get said advertising. It is also, probably, criminal.
Apple appears to have made a deliberate corporate decision to be incompatible with OpenOffice.
I disagree with this. Rather, I suspect Apple has not prioritized ODF and created their iWork applications based upon their own format for convenience. If Apple as a company was trying to be incompatible with OpenOffice they would not have added ODF support to TextEdit. I would venture to guess that Apple sees the business case for supporting import and export to MSOffice formats, but is as yet unconvinced about ODF and whether it will succeed in the marketplace. I hope that they understand that they can help to influence the direction of the market in a positive way by moving to ODF as the native format for their offerings, but even I am not sure that is really the case.
There is really nothing to be gained by Apple for helping open standards; they'll probably just license OOXML and tout their Windows compatibility.
Actually there is (potentially) something to be gained. By supporting ODF they can become one more vendor that helps demonstrate the benefit of ODF over OOXML. They can also qualify as a vendor for purchases in the future that require ODF (as some government agencies are now moving towards). Anything that hurts Microsoft and weakens their monopoly influences, leads to a better market for Apple to compete in. The only question for Apple is if it is worth the cost and is the battle already lost by ODF?
I really think hybrid applications that are both traditional and Web apps are going to be the way of the future. Local applications don't allow you to edit from any machine, are not automatically kept up to date (payware), and don't allow developers to easily leverage ad revenue or subscriptions. They are not as simple for collaboration and publishing to the Web. They are not as easily targeted to all platforms because of lack of standardization for running applications across OS's
Web apps are reliant upon a network connection, don't provide the security demanded by some use cases, and are not good at finding geographically close users. Performance is limited by network throughput and latency.
Really in a free market the direction of development is almost certain to go to apps that connect to internet services or apps that are also internet servers. They both come down to the same thing, just differing in the emphasis on decentralization or centralization. Given that the network is the more common limiting factor today (especially in the US and the third world) hybrid apps like this offering are probably going to be very big, very soon. The only thing holding this back has been Microsoft's ability to cripple Web technologies and their monopoly influence in the office suite market.
I'm always saddened when people who have obviously not really used both OS's being discussed try to argue the superiority of one over the other. Seriously, don't even bother unless you actually know both OS's and what they can do.
Sonnet - It isn't simply spell check but a bevy of language tools including language recognition and grammar check.
Yes, it is an API as well. It is also still worlds behind OS X's very innovative system services which has allowed users to plug in arbitrary functionality via a simple plug-in or offer functionality contained within an application. With Sonnet, you can add functionality like grammar checking and application developers can then alter their programs to take advantage of it. With OS X, all cocoa apps can automatically use these functions, even if the developer never even considered a use. I dropped a grammar checking service into my Services folder in 2001 and have been able to use it in Safari, Terminal, MS Word, TextEdit, Mail.app, iChat, etc. ever since. I also dropped in some text manipulation scripts, language translation services, automatic bibliography formatting, etc. When I installed a payware professional thesaurus it offered itself for use in all the other programs as well. KDE is ahead of OS X in a number of ways, but this definitely is not one of them.
Actually OS X does have tools to do the same sort of thing, but they don't ship with a GUI, although there are several freeware GUIs. That said, I hope Nepomuk helps to bring this sort of functionality into the mainstream with more ubiquitous, standardized use across program UIs, in a default install.
Solid - I don't know anything about I/OKit, but Solid provides an easy API for apps to interact with hardware.
I've never used Solid, but I don't see anything to imply it is revolutionary, or even a best of breed solution.
Plasma - In many ways it offers previous functionality (panels, dock, widgets, etc) but it brings them all together under one library and framework, however the real innovation and advantage of Plasma is the ability to generate apps easily in most any language.
Even Windows Vista offers this these days.
And while widgets have existed for a while, most widgets are useless toys. Plasma provides powerful data engines to create actual useful apps.
There are certainly some very useful Widgets and it is nice that KDE can use ones designed for OS X (even if I will probably never use them). Nothing prevents Widgets from having the functionality of full fledged apps and on OS X they can plug into the normal APIs like CoreData.
For quite some time I've been maintaining a large list of the advantages each major OS has over competitors. Maybe I should expand that into a full-fledged comparison and submit the article to Slashdot. Then at least users can have an informed list they can use as a basis for comparisons. It just seems like most arguments turn into emotionally charged messes with very few people arguing from the position of actually having used on a regular basis the OS's in question. The only thing holding me back from writing it right now is that I've ditched Vista for the time being and am on WinXP for my Windows needs (until Vista stabilizes).
If you want to see where OS X has borrowed from OSS, simply look at spaces (predated by Virtual Desktops), Dashboard (predated by Konfabulator), Spotlight (predated by Beagle), etc.
I think it is pretty clear that Apple has been listening to their user base, many of who are old Linux/UNIX users (as well as a good chunk of new employees) and implementing things people miss from other platforms. Apple has also done some serious innovating in OS X: expose, ubiquitous zeroconf, system services, launchd, etc. What I find interesting is that major Linux developers don't copy those features from OS X. I have several theories as to why this is.
It may be that most of these features are useful for a desktop OS, and most linux developers are focused on Linux as a server and, in fact, will try to block anything not useful for use as a server because for their purpose it just adds bloat and/or instability.
Linux desktop developers may have never used OS X and are unaware of what these features are or how they work. Many Linux users who do use OS X, seem to become switchers and no longer need to scratch that itch since they get the features just by using OS X.
Linux desktop developers have a Not Invented Here syndrome and are not willing to even consider technologies from Apple, instead concentrating only on Windows as an alternative they need to keep up with.
Many of these features are major architectural changes and/or are not backwards compatible and it may be that no one Linux distro has enough momentum that it can make such major moves that might make it less compatible with other versions of Linux and attempts at advocating such standards via groups that span distros is simply too complicated. Linux has no Steve Jobs to make a decision and push it through (which is very good in some cases and very bad in others).
I'd like to see some OS X fanboys who have a clue about the way OS development works; hint: all the majors copy concepts from each other & rarely come up with original features (they mostly come from research projects)
Some of those research projects are internal to a given OS vendor. In any case, there is a difference between copying a unproved new technology from a research project and copying a mainstream feature in use from another major OS. The latter takes less effort and is less risky.
I kinda wish adobe would make a decent pdf viewer for linux, or at least make the pdf format open
Umm, PDF is an open standard, one with many, many implementations. Adobe's reader is not terrible, but its default configuration is (many slow and bloated plugins). Alternately you can use Foxit on Windows, Preview on OS X, XPDF of KPDF on Linux. There are actually quite a few more readers if you want to try others. You can make PDFs using any application on OS X, using CUPs on Linux, MS Office, OpenOffice, Wordperfect, Docbook, etc.
It always tells me "Express Install not supported by this version of the Flash Player..." But I have the latest version of Flash for Linux installed...
There is a bug in several Linux distros (including Ubuntu) where installing the latest Flash fails silently and tells you you have the most recent when you do not. You need to install by hand. In any case, it does work if you have everything installed properly. I tested it with Flash Player 9.0.115, Firefox 3Beta4, and Ubuntu 8.04.
Only an engineer would be worried over the "problem" of how to allow people in a coffee shop to talk to one another.
I didn't say "talk" I said "chat." On Slashdot I should think most people would understand the difference. Reading aloud the characters in a URL you want everyone to visit is a lot less efficient than just sending a link to everyone via IM. Ditto for sending a file to everyone, or really any other local networking.
I am worried that Apple is assuming too much about the security of the Mac OS X operating system.
I agree with this. From what I've seen it seems like there are several internal developer types at Apple:
Classic Apple employees - they've been with Apple for a long time. They know UIs, but have never really focused on security and don't care about it.
Next employees - they came with Apple's acquisition and have some security savvy, but it is not a major concern.
UNIX/Linux/BSD geeks - either hired on in recent years or acquired with Apple's small acquisitions, these are old school guys who consider security as part of everything they do. They are probably responsible for a lot of the security in Apple products along with the efforts of other OSS contributors.
It seems like at some point Apple needs to step up and form a real security penetration testing team, that will spend all their time trying to hack OS X and Apple products. Heck hire Charlie Miller and pay him to keep on doing this. Give him some stock options and make use of his expertise.
I'd really like for Apple to "step up" and take this bull by the horns and let the world know that they are very serious about security and eliminating *any* means of intrusion, either automated or user driven... and not just rely on the FOSS community to remedy the security problems in the software that they have incorporated into the OS.
Actually, Apple has been doing some serious work for advancing security. Leopard has a mandatory access control framework built in, that can be used to sandbox applications. They even sandbox a few services by default adding another layer of security, just as vista does. They do not, however, sandbox Safari by default. Leopard also has an application signing framework, that can be used to help authenticate code origins, an important area for the direction the security industry is taking.
Just as long as they don't implement some Vista like "Allow or Deny?" crap... God that would drive me *nuts*!
This security "feature" in Vista is MS's usual user interface design nightmare. They can't seem to wrap their heads around UI design as part of security and seem focused on making sure the user can be blamed instead of just trying to reduce the chances of an exploit working. I would love it if Apple brought restrictive, default MACLs to Leopard, but did the UI and MACL generation properly. You shouldn't be asking users over and over to click Allow or Deny and thereby conditioning them to click Allow to "make stuff work. MACLs should come from Apple and other security vendors. Users should be asked who they trust (Apple, ClamAV, Symantec, Norton, etc. ). Applications should be restricted according to MACLs that ship from the software creator and from security vendors based upon user preferences. Adding security should be easy as buying a subscription to Symantec's MACLs and deciding how much weight to give them. Symantec should install no software... just supply data in an open and defined protocol. Users should almost never be asked to allow or deny a given behavior from an application and only then if it is non-mainstream software and it is trying to do something very unusual - kernel patches, UI hacks, and malware should be about all they are asked about and then the question should be informative with explicit verbs for button, eg, "Program 'Martian Hunter' wants to access your address book (Stop it from accessing your contacts)(allow it to access your contacts once)(allow it to access your contacts always)." Further, whenever possible the OS should try running applications denied access without granting privileges and if that fails run them handing over bogus data and if that too fails try running them in a VM with bogus data and bogus access to other resources before failing entirely. In this way users don't associate granting access or not granting access with whether or not they can use an application. Users want to run stupid little applications and the OS needs to allow them to do so (even if they are malware) but in as safe a fashion as possible.
Security is all about layers. Vista's sandbox model for IE is another security layer that Safair is lacking. The anti-phishing features in IE and other browsers are another are another layer. None of the layers are perfect, but they stop a class of attacks.
I agree with your opinion on this. Safari+OS X is lacking several security mechanism in use on Vista+IE by default. Firefox+Ubuntu is likewise lacking one layer. That said, in many cases there are performance and ease of use tradeoffs with implementing additional layers of security. Further, this contest was about targeted attacks by "crackers" which is an uncommon situation for normal consumers.
Because of our malware ecosystem, I'd argue that the appropriate level of security is different for different operating systems and other software and use cases. Normal users don't worry about attacks by a dedicated individual going after just them, but rather about automated worms and widespread malware and phishing. Apple did not lock down Safari in a sandbox. They did not include their anti-phishing feature in a regular release of Safari. I would rather they had done so. But it is not a real concern for most users simply because malware and even phishing is not a major concern to most OS X+Safari users. The risk of a worm or regular phishing attack is very small, so small in fact that to date installing antivirus software that might malfunction has proved the greater danger. While I don't know about the number of phishing attacks, I'd argue that the risk of being compromised is greater if a user switches to Firefox, because there have been in the wild malware attacking Firefox, whereas that is not the case with Safari to date. I'd further argue that it is probable that sandboxing Safari in OS X would have caused more problems for users (due to configuration issues, or malfunctioning sandboxes) than releasing it. Likewise for sandboxing Firefox by default on Ubuntu.
A lot of people complain about Microsoft's security problems with Windows. I think they have a very valid point. A lot of people tout the security features in Windows. They have a very valid point as well. The disconnect and apparent contradiction is explained by looking at the context and how it affects the average user. Windows has not, to date, had an appropriate level of security such that it is not an issue for the average user. Both XP and Vista have suffered from failure to stop "in the wild" automated attacks that caused problems for users. Both have also suffered because misconfigured or malfunctioning security features have caused problems for average users. Microsoft has a lot of work to do in both of these areas.
Both OS X and Ubuntu to date have not had significant problems for normal users due to security failures or malfunctioning security features. This is to their credit, but has more to do with delivering the correct level of security to suit their users (both by default and hardened versions for higher risk deployments). That should not be confused with claims that they have more or better implemented security features. If MS's monopoly market share ever should fall, both those OS's will have a lot of work to do... but of course everyone would also have direct financial interest in making those improvements.
I got the same behavior on two different Linux machines, Gentoo and Fedora Core, one running Firefox 2.0.0.4 and the other 2.0.0.12. I tried re-installing Flash 9, didn't make any difference.
Well my configuration is certainly different. I have Flash Player 9.0.115, Firefox 3Beta4, and Ubuntu 8.04. I had to manually install the Flash player because of a known bug. Apparently Adobe released a new version of Flash 9 in December and on many systems the install checks the MD5 checksum, sees it is different, and silently fails while reporting that it was installed. You might want to try installing Flash by hand or just trying this site again in May, after all the distros manage to patch this. I would argue that since it works for me on Linux, your problem is probably a bug in the version of Firefox or is an issue of an outdated version of Flash combined with a bug in your package management.
This certainly does bring up a long standing complaint I have with every desktop Linux distribution I've tried. Linux is in the lead as far as package managers are concerned... and with ease of finding and installing open source, freeware. At the same time, package managers on Linux are terrible when it comes to commercial software, especially shareware/payware and behind the curve for software portability. I really wish Redhat, Sun, IBM, Canonical, and Apple would all sit down together and create a standard package format based upon GNUStep, and define a set of protocols needed for discovery, transfer, compilation, registration, licensing, installation, and uninstallation of software. I fear, however, that Linux's use as a server and appliance will prevent any such advancement under the auspices of preventing "bloat" with an unhealthy dose of Not Invented Here Syndrome.
I received my XO about a week ago, live in the US, and I can't connect it to the internet reliably. Googling around, I found a lot of discussion about how poorly the XO laptops work with wireless access points.
The XO was designed and tested to work in conjunction with other XO laptops and with the XO server. Connecting to generic wireless in the states is nice, but you must admit not at all what it was designed for.
This problem certainly took some shine off an otherwise cool little computer for kids.
I've heard a lot of stories about people buying XO laptops for their own kid or to play with. I certainly don't have a problem with that, but I do worry that a lot of opinions about the XO will be formed from this type of use, which is not the use for which it was designed. If all the kids in a group don't have one and a server to connect to periodically, well you've just lost a huge chunk of the functionality.
The American kid with a Windows PC or a cell phone doesn't seem to having much trouble networking with anyone.
If only that were true. Do you know how much work it is to start a group chat with everyone in say, the coffee shop? Windows auto-discovery of local services/users sucks badly, especially compared to the XO laptop.
"Any" operating system? Not Linux! I can't even open this damn "program". I get a message that says: "Express install is not supported by this version of the Flash Player..."
And you didn't think upgrading your Flash player would be a good idea? It works fine for me with Firefox and Flash 9 under Ubuntu. (Mind you Flash installation, or really any commercial software, could be a lot easier.)
The Manhattan Project was top secret because we didn't want the enemy to get the atom bomb before we did.
It's called "humor." You're right, of course. Also we didn't want them to know we were working on such a device for other reasons (sabotage, etc.). I actually used to know a brilliant mathematician who had worked on the project. Sadly he was suffering from Alzheimer's disease and while he could still remember all the trig tables, he occasionally would forget where he was or who he was talking to. Still, he had some great stories. PBS did a good series on the project as well. Requisitioning gold from Fort Knox to use for wiring was my favorite bit of trivia.
After reading the tenth or twentieth scientific article that interviewed people working on the LHC, that includes some wild speculation about remote possibilities that might come to pass when it comes online... this surprises me not at all. I understand being a bit sensationalist to make a more entertaining article. I understand hyping the potential a bit to help keep that government funding coming in. Still, black holes, strangelets, cascading subatomic events, time travelers finding the earliest point to return to... it was a bit much. Maybe you get promoted in experimental physics by making waves and smoking pot with the boss. The you want your name in a magazine so you spin some half-assed idea as though it was a real possibility. The only problem is, some people listened and are now worried.
This is why the Manhattan project was top-secret: two out of six physicists think it might destroy the planet... okay those are good odds, let's try it.
What many of these kids won't get is Internet access: about 90 percent of the villages lack it, and may not get it anytime soon.'"
This is a serious drawback as the internet is a great source of information as well as a way to commercialize the computing efforts of these kids and (potentially) give them some good old capitalist reasons to study hard. Even so, while the lack of internet access is not as big of a drawback as it might be. These laptops will presumably still form their own mesh networks and connect to the school's XO server. I bet a lot of Windows using American kids wish their computers would allow them to network with friends nearly as easily.
Best of luck to these teachers. It is always hard being the guinea pig for a new technology and it will probably take a lot of dedication to alter their teaching methods to really take advantage of this new tool.
Slashdot Mirror
Memory is a big deal for me because I have a cheap laptop with as much RAM as it can hold. Still RAM can be the limiting factor for two reasons: First, I often have another OS running in a VM, which really puts a dent in your available RAM. Second, My OS and most user software have stabilized and suspend modes have become fast enough. Things just don't crash anymore and as a result I just don't reboot anymore. My uptime on my laptop is generally the same as a few days after the last security patch was released. Sometimes this is months.
Really I don't care if Firefox uses twice as much memory as it does during the first hour I use it. The problem comes when I open a few tabs of reference material, then leave those tabs open in the background until I get back to them. In this instance I've gone back to Firefox to kill it, because it gradually used more and more RAM until other applications were starved. The only other programs I've used that exhibited that behavior are Adobe InDesign (which is even worse than Firefox, 2 days between restarting it, tops) and MS Word (which is about the same as Firefox.
I mean Firefox has had some nasty memory leaks for the longest time and absolutely I would love to see those fixed. But it seems like this is more than just that, it seems like some big epeen contest between browsers.I know it is uncommon in the computing industry these days, but it is called 'competition'. There are several viable alternatives for your browser of choice and each is striving to outdo the others and gain market share. Sadly, this competition does not extend to Microsoft who is still abusing their monopoly to take most of the market. The end result, however, is that among other browsers a lot of the market is savvy enough to see what memory usage is easily and hence use it as a benchmark of sorts.
...or one like IE 7 which is more polished than both Firefox and Opera.Heh, your milage many vary, caveat emptor, this statement does not reflect the views of this station or geekdom, etc.
But I don't hate MS. I do hate what they've done to the science of computing. I don't want them to fail as a company, but I do want them to fail at their illegal business practices.
...and you're[sic] zealotry wants ODF to succeed...I want ODF to succeed and I want the current version of OOXML to fail, because I want the office suite market to be competitive and to actually advance. I want Microsoft to have to make MS Office the best offering on the market and keep improving it to keep it the best if they want to have a large market share. I have no problem with MS Office being the most widely used office suite, but I want MS to have to earn that status by making it better than everything else, not by breaking compatibility with others and making it hard for users to use anything else because of their shenanigans.
...and that you're worried that now the OOXML is a standardNo, now I'm worried that purchasing agents will think it is a standard and will accept it as if it was a standard, not realizing it won't bring them the advantages a real standard like ODF can.
...the huge install base will put a lot more implementions of OOXML out there than ODF.Yes that is exactly right. MS's huge market share will put a lot of implementations out there compared to ODF. The fact that you don't recognize leveraging monopoly influence as a bad thing is very sad. I have no problem with OOXML being implemented more than ODF, but it should be because it is a better standard, not because Microsoft is pushing it on people who generally don't have the power to do anything about it.
Kompose is only one of the expose-type features on linux. Another is the expose-like feature in Compiz, enabled by default in Ubuntu. Kubuntu is a community-supported distro & doesn't develop as quickly as Ubuntu.
Really, you can activate an expose like feature in a default Ubuntu install. Do tell me how, since I recently switched to Ubuntu as my main Linux desktop. What mouse button do you press? What key combo do you press? I saw no mention of this anywhere in the help system and have not yet got around to searching for a 3rd party solution.
So - according to you, 'linux' developers aren't interested in zeroconf, but you even point out in your own post that the XO linux developers do support ubiquitous zeroconf.
XO is not a mainstream Linux distro. None of the major Linux distros has made ubiquitous use of zeroconf. Sure, a few developers on a special project aimed at kids did. Everyone else has largely ignored it. Why isn't it used in Ubuntu and Fedora and Suse? Because most Linux on the desktop developers did not bother to copy OS X in this regard. That is a serious failing. Unless you're suggesting that the XO laptop constitutes a viable choice for me to use as my desktop, my point stands.
I misread that as SystemStartup - my apologies, linux indeed does not have something like system services - this is the consequence of being a bazaar rather than cathedral project.
As I said, there is no one person who can push major changes, so in general they don't happen. This is both a strength and weakness for Linux. Pointing out where it is a weakness should not require me to go through several revisions of comments with people who claim to know enough to compare said strengths and weaknesses.
Oooookay? Again - you said "What I find interesting is that major Linux developers don't copy those features from OS X." - not "What I find interesting is that major Linux developers don't copy those features from OS X - and instantly have working implementations"
The point is, while someone is "working on it" for many major distributions, we have no idea if those projects will ever be integrated into the mainstream distributions. There are 10 projects on sourceforge for every one that becomes usable. It has been 3 years since Apple shipped a working version of LaunchD and made the code available. None of the major Linux distros has managed to copy or create their own and ship it in their stable release. At this point, I have my doubts that any will for a long time.
So they're trying to implement it, but thats still not good enough for you.
That is correct. I want there to be enough linux developers familiar with OS X features such that they realize the advantages of those features and prioritize them. Sadly, I see just the opposite. I know a lot of Linux developers, but a good portion of them now use OS X as their desktop and actually oppose cloning OS X features in Linux, not because they aren't useful (they use them daily) but because they aren't useful for a server, which is what they want to use Linux for.
Recent OS X switchers tend to come from Windows via Linux (for a month or two) & don't really have 'real experience' with Linux.
Yeah, that is not at all my experience. I've spent the last 8 years working in various companies that make heavy use of Linux and *BSD in making servers and routers. 8 years ago I knew one or two Mac users. Last year I left a small company of a few hundred developers and more than half were using OS X as their desktop to develop Linux and software that ran on top of Linux. All of them had significant experience with Linux or we would never have hired them.
Furthermore, if we're swapping anecdotes, the closest I've come to people with real experience using both are people who actually use both daily. I'm one of them (along with windows).
Gee that's gr
Sigh. Why is explaining deficiencies in Linux like pulling teeth? Seriously, why are Linux users in general so opposed to understanding ways someone else has done better and trying to improve based upon them? I'll go through your bullet points one by one.
Zeroconf - Avahi
Yes, there are certainly zeroconf implementations on Linux. My point was "ubiquitous zeroconf." You now what 'ubiquitous' means, right? Both Pidgin and Kopete support zeroconf for discovering local chat clients. Aside from that and a general zeroconf browser in Ubuntu, they don't make use of zeroconf for much of anything else. They don't use it to discover terminal servers or remote GUI sessions or shared streaming music. Niether Ubuntu or Kubuntu finds my printer, despite it advertising itself via zeroconf. About the only Linux distro I know of that makes what I'd refer to as ubiquitous use of zeroconf is the XO laptop. Seriously, all major Linux distros I've used have ignored the potential uses. I don't even know of a multi-user text editor that uses zeroconf on Linux. For servers, I can see firewalling it off by default, but for the average desktop, on the local network, it makes sense for the huge functionality win.
Expose - Kompose
Yes, somebody cloned Expose for KDE. It even works fairly well (at least on my Kubuntu box). I still had to go out and search for it and then add the package by hand. That simply isn't going to happen for most Linux newbies, or even Linux users who haven't seen it on OS X. It is nice that I can add such a thing, but this does nothing to convince me the Linux on the desktop developers are willing to adopt real innovations from OS X. Ask me again when it is enabled by default on a major distro.
System services - errr, that's been replaced by Launchd on your list.
You see. This is exactly what I'm talking about. You don't even know what "system services" are on OS X. How can I believe you are making an informed comparison, when you obviously have never taken the time to learn what OS X's real strengths are. I've had this same conversation with a baker's dozen Linux on the desktop developers. None of them knew what this feature was and once explained they all agreed it would be useful... but too much work to try by themselves.
System services on OS X allow functionality to be offered by plug-ins and applications to other applications, universally, without the developer of the other application doing any additional work. It is how OS X offers system wide functions like spell checking and grammar checking, but also allows the user to add arbitrary functionality on par with those. For example, I long ago added a service to translate between german and english, and now it is a key combo away in my mail client, chat client, and web browser. When I installed Graphviz it offered its service to other applications so in any other program (terminal, textedit, excel) I can just highlight data and hit a key combo and it sicks it into Graphviz and creates a graph of the data, which it loads into my copy/paste buffer. I don't have to bother wasting time opening the program and copying and pasting data and rendering it. I probably use a dozen different system services in my average work day for spelling, grammar, bibliography auto-formatting, dictionary/thesaurus/wikipedia lookup, text manipulation scripts, etc. When I did development work more, I used a different set of them. The whole point is it allows the user to customize the functionality in programs, without having to write any code and recompile, or even have the source to a given application. No Linux distro I've tried comes close and no KParts is not "just as good" or even the same concept.
Launchd - ReplacementInit - Considered Apple's LaunchD and Sun's SMF amongst others, but none were quite the right fit.
Yes, every distro has such a project and as far as I know none of them are compatible (except as backward compatible mode
Which "they" to you mean here? ISO (and to a lesser extent, Ecma) would be the proper "they" in this case - not Microsoft - and you never meant to imply otherwise, I hope.
Actually I did mean to imply MS should have provided a version of the spec that solves all the major problems noted with it by others.
Incorporating all of the BRM recommended changes into a usable standard is something that's being worked on right now, and will continue if DIS2900 has in fact passed this vote.
That's an interesting theory. I take it you've never heard of Microsoft before today?
It has already been agreed that the parts which reference Word97 (which were always clearly marked as deprecated and optional) will be moved to a new section - also clearly marked 'deprecated and optional'.
That's not good enough. If it is included in the spec even as an optional component, MS can implement it and their documents can break when other programs that also conform to the spec (but don't implement optional portions) open them. This undermines the point of having a standard.
I guess you're going by the SFLC's recent remarks about the OSP here. Oddly they didn't feel the same way about remarkably similar covenants from IBM and Sun.
That's not odd at all. With Sun and IBM it is academic because neither has monopoly influence on the market and as such can't undermine competition by those means. MS does have monopoly influence in the office suite market (although they've spent a buttload of money settling lawsuits trying to make sure no court makes that declaration). When a document breaks when written by IBM's software, people can avoid using IBM's software or users can complain and IBM will fix it. When the same thing happens with documents saved by MS Office, because of its overwhelming market share, it doesn't matter who caused the document to break, it will always be treated by the market as a flaw in the program that is not from MS, thus introducing artificial problems in competitors and undermining the operation of the capitalist free market.
There would seem to be some dissent within SFLC as well, given that Mark Webbink (of SFLC's board, and also speaking as general counsel for Red Hat) has said that he feels the OSP does indeed allow the flexibility to be implemented under F/OSS licenses.
Of course there is dissent, that is why there is a voting process. It is just that the voting process was not designed to be properly robust in the face of a big company with a lot of money and influence trying to undermine the process.
If you've been involved with the writing of some ISO standards, and you're an advocate of free and open standards, surely it rankles you that many ISO standards aren't freely available, but instead must be purchased for hundreds of dollars ($350 or so for ISO/IEC 26300, a/k/a ODF)?
Not really. Open source or even Free (as in freedom) is not the same thing as free (as in beer). I'm an advocate of open standards because they allow free markets to operate efficiently. That is efficiently sans intervention by an abusive monopolist which has enough political influence and legal savvy to render the courts unable to effectively stop its abuse.
But I digress (sorry) - I have personally looked at both ODF and OOXML. Haven't just taken the pundits opinions on faith - I went and looked. Both have large sections which require much re-reading if one wants to create a good implementation.
Did you perhaps notice that every function in OOXML simply describes how MS Office or the largest third party vendor that interoperates does things. Did you perhaps, notice that the format eschews consistency within itself for the sake of more closely describing what the various component programs of MS Office does? ODF is not perfect, but it sure doesn't take a genius to see the vastly different mindsets appli
While numbers are hard to calculate for OSS desktop OS's, I haven't seen numbers to support your assertion. That isn't too important though because you are misunderstanding the market distinctions. Apple doesn't sell desktop OS's. They use vertical bundling to bypass the desktop OS market and instead compete in the computer system market. OSS OS's have a very small market share pre-installed or even OS-less whiteboxes.
What does that have to do with anything?It speaks to the viability of NeoOffice in professional use as a competitor to Apple's office suite. I've seen it there but not on a large scale, not even on the scale of iWork (let alone MS Office).
Well, hey, no need to buy iWork then, right? Don't make me laugh.Umm, I don't see what you're trying to get at. You argued that Apple as a company was opposed to letting ODF gain a foothold. When I countered with Apple's support in TextEdit you claimed that maybe they only wanted users to be able to read the occasional ODF document, but at the same time keep it from being viable as a competitor. Apple's support for writing ODF contradicts that theory. Claiming now that it isn't important on the third iteration of this subject is weak. Just concede the point gracefully and move on.
Yes, and they could have adopted OpenOffice XML at the timeNo, they couldn't have since it wasn't finished. They could have moved to a draft of it and tried to influence the spec enough to work for iWork and OpenOffice, but there is no guarantee they would have been successful. Of course at that point the developers may not have even heard of it let alone considered it viable. We don't even know if they started from scratch with iWork or used an existing code base from Apple or Next. In short, Apple could have considered ODF and possibly adopted it, but we have no idea if they evaluated it or if they did, how risky they considered such a move or how much more expensive it would have been for them. Remember, at that time the spec was not implemented by anyone and there was no guarantee it would be. Further, the spec was at that point completely under Sun's control. For all they knew at that point it was never going to be implemented by anyone else.
Instead, they went through two poorly designed XML formats that were homegrown.That's an interesting assertion. How, exactly, do you find Apple's XML formats to be poorly designed (and for that matter which formats are you referring to, iWork has 3 formats: .key, .pages, and .numbers).
Apple has a long history of intentional incompatibilitiesThat's very interesting, but what does that have to do with whether or not they are intentionally making iWork incompatible? At the time of Apple's creation of iWork, there was no approved open standard for office suites and not even OpenOffice would support ODF (when it was finished) for another year.
Apple has no interest in helping ODF adoption or risking their relationship with Microsoft.Great two more assertions. You do know you have to do more than make assertions in order to make a persuasive argument, right? How about backing up some of your assertions with some facts or logic. Everything you've presented so far has been an empty assertion or a fact I have demonstrated is false or does not prove what you hoped it did. In logic, facts precede forming an opinion. Why don't schools teach logic anymore?
Apple doesn't really compete with OSS in the desktop space. It might in the future, but to date OSS does not have enough market share to count. Apple does compete with Microsoft and with Microsoft's customers. Microsoft has been leveraging their office suite to gain an advantage in other markets. Ceding the office suite market to MS hurts Apple's iWork sales. It hurts their computer sales too. MS having monopoly influence in the office suite market, but not leveraging it directly against Apple in the computer market is where Apple is now. They know they would be in a better position yet if they could get rid of MS's monopoly influence in the office suite market, especially since MS has announced changes that will lead to their exploiting it against apple (dropping vb scripts in office for the mac).
They can do that with NeoOffice.Currently, I wouldn't recommend it. NeoOffice is just too slow and buggy for large scale production use (IMHO). In any case, that only applies to sales of computer systems, not to sales of office suites. Apple sells an office suite called iWork, which currently is not in the running for those contracts.
Makes perfect sense: they want people to be able to read the occasional ODF file, but they don't want people to strengthen ODF by fully supporting it.Except TextEdit has both read and write support for ODF. If they wanted to keep ODF from competing with them, why would they provide support for writing to it? Also, TextEdit ships with every OS X system, iWork does not. That means OS X itself has better support for ODF, by default, than Apple's own iWork formats.
Well, you can "disagree" all you want, but it isn't born out by the facts.I think it is.
Apple has spent a lot of time and money developing two different XML formats for iWork when they could have just adopted the ODF format.The first reports of iWork in development were in 2003. It was officially announced in 2004 and was actually available in 2005. ODF 1.0 was finalized in 2006. Draft versions of ODF existed as early as 2002. For reference OpenOffice.org did not support ODF until 2005, the same year iWork was already on sale. It did not gain momentum among any other vendors (except Sun) for another year.
If the iWork team at Apple were aware of the upcoming ODF standard, they could have joined the group and helped create the first version (and used it in Pages). As it was though, Pages was not being targeted as a Word replacement for general word processing. Rather it was targeted as a layout application for home users, sort of like InDesign for home users. Many of the features it used were targeted at that market segment and were quite different and even if Apple had joined there is no guarantee they would have had enough influence to make ODF appropriate for their target market. Even today the committee is pretty much Sun and IBM employees. There are a few Novell people and Chinese government people, two guys from Adobe, and one guy from Google. That's pretty much it. It would be quite nice if Apple joined and provided support in their applications. They may, eventually, do that, especially if the formats see wider uptake in the future. Claiming that Apple is intentionally avoiding compatibility is a stretch. It is probably mostly a matter of which types of employees have influence on the iWork project. It is probably old school Apple guys (who tend to largely ignore standards and security) as opposed to former Next employees and people hired in from the Linux/BSD/Unix community.
No one, not even MS can implement it because they haven't published a version that makes the required changes, it is ambiguous, it references other formats that are not published, and it contradicts itself. Basically, it is just like the existing MS Office formats. You can't follow a standard. You just have to try to reverse engineer what MS did for the common cases in the most popular version of MSOffice and ignore the uncommon cases.
So what? I don't recall ISO standards as having to be free of patents.In the Office suite industry two of the four biggest competitors are OSS projects. Creating standards that for legal reasons are not implementable by such a huge portion of the market, undermines the standards process. This is MS's normal tactic, copy a feature of other software, but at the same time undermine the benefits that feature brings to users. It is creating a hollow bullet point for salespeople while not benefiting users in a way that might allow them to choose anything other than MS.
Have you read other ISO standards?I have. I've even been involved in writing some of them. I dare you to go look at ODF and then look at OOXML. The difference is night and day, even for a layperson.
If it were approved, it would be a standard..No it wouldn't. Some committee rubber stamping something does not make it a standard. ISO's purpose is not to make an existing format a standard, but to help create and certify that formats are standards. By them certifying OOXML as a standard does not make it one; it just fools people who have relied upon ISO's reputation into thinking it is one. OOXML will not be any more usable or implementable by others or clearer next week after ISO announces they have approved it than it is today. It is just false advertising bought by Microsoft. Strangely some people here object to false advertising, especially when MS has destroyed the credibility and integrity of a standards body to get said advertising. It is also, probably, criminal.
I disagree with this. Rather, I suspect Apple has not prioritized ODF and created their iWork applications based upon their own format for convenience. If Apple as a company was trying to be incompatible with OpenOffice they would not have added ODF support to TextEdit. I would venture to guess that Apple sees the business case for supporting import and export to MSOffice formats, but is as yet unconvinced about ODF and whether it will succeed in the marketplace. I hope that they understand that they can help to influence the direction of the market in a positive way by moving to ODF as the native format for their offerings, but even I am not sure that is really the case.
There is really nothing to be gained by Apple for helping open standards; they'll probably just license OOXML and tout their Windows compatibility.Actually there is (potentially) something to be gained. By supporting ODF they can become one more vendor that helps demonstrate the benefit of ODF over OOXML. They can also qualify as a vendor for purchases in the future that require ODF (as some government agencies are now moving towards). Anything that hurts Microsoft and weakens their monopoly influences, leads to a better market for Apple to compete in. The only question for Apple is if it is worth the cost and is the battle already lost by ODF?
I really think hybrid applications that are both traditional and Web apps are going to be the way of the future. Local applications don't allow you to edit from any machine, are not automatically kept up to date (payware), and don't allow developers to easily leverage ad revenue or subscriptions. They are not as simple for collaboration and publishing to the Web. They are not as easily targeted to all platforms because of lack of standardization for running applications across OS's
Web apps are reliant upon a network connection, don't provide the security demanded by some use cases, and are not good at finding geographically close users. Performance is limited by network throughput and latency.
Really in a free market the direction of development is almost certain to go to apps that connect to internet services or apps that are also internet servers. They both come down to the same thing, just differing in the emphasis on decentralization or centralization. Given that the network is the more common limiting factor today (especially in the US and the third world) hybrid apps like this offering are probably going to be very big, very soon. The only thing holding this back has been Microsoft's ability to cripple Web technologies and their monopoly influence in the office suite market.
Lets see, the highest market share distributions using Gnome are:
It looks to me like there are as many KDE users as Gnome and it doesn't really correlate to where the company is based.
I'm always saddened when people who have obviously not really used both OS's being discussed try to argue the superiority of one over the other. Seriously, don't even bother unless you actually know both OS's and what they can do.
Sonnet - It isn't simply spell check but a bevy of language tools including language recognition and grammar check.Yes, it is an API as well. It is also still worlds behind OS X's very innovative system services which has allowed users to plug in arbitrary functionality via a simple plug-in or offer functionality contained within an application. With Sonnet, you can add functionality like grammar checking and application developers can then alter their programs to take advantage of it. With OS X, all cocoa apps can automatically use these functions, even if the developer never even considered a use. I dropped a grammar checking service into my Services folder in 2001 and have been able to use it in Safari, Terminal, MS Word, TextEdit, Mail.app, iChat, etc. ever since. I also dropped in some text manipulation scripts, language translation services, automatic bibliography formatting, etc. When I installed a payware professional thesaurus it offered itself for use in all the other programs as well. KDE is ahead of OS X in a number of ways, but this definitely is not one of them.
Nepomuk - This is FAR more than search. http://nepomuk.kde.org/Actually OS X does have tools to do the same sort of thing, but they don't ship with a GUI, although there are several freeware GUIs. That said, I hope Nepomuk helps to bring this sort of functionality into the mainstream with more ubiquitous, standardized use across program UIs, in a default install.
Solid - I don't know anything about I/OKit, but Solid provides an easy API for apps to interact with hardware.I've never used Solid, but I don't see anything to imply it is revolutionary, or even a best of breed solution.
Plasma - In many ways it offers previous functionality (panels, dock, widgets, etc) but it brings them all together under one library and framework, however the real innovation and advantage of Plasma is the ability to generate apps easily in most any language.Even Windows Vista offers this these days.
And while widgets have existed for a while, most widgets are useless toys. Plasma provides powerful data engines to create actual useful apps.There are certainly some very useful Widgets and it is nice that KDE can use ones designed for OS X (even if I will probably never use them). Nothing prevents Widgets from having the functionality of full fledged apps and on OS X they can plug into the normal APIs like CoreData.
For quite some time I've been maintaining a large list of the advantages each major OS has over competitors. Maybe I should expand that into a full-fledged comparison and submit the article to Slashdot. Then at least users can have an informed list they can use as a basis for comparisons. It just seems like most arguments turn into emotionally charged messes with very few people arguing from the position of actually having used on a regular basis the OS's in question. The only thing holding me back from writing it right now is that I've ditched Vista for the time being and am on WinXP for my Windows needs (until Vista stabilizes).
I think it is pretty clear that Apple has been listening to their user base, many of who are old Linux/UNIX users (as well as a good chunk of new employees) and implementing things people miss from other platforms. Apple has also done some serious innovating in OS X: expose, ubiquitous zeroconf, system services, launchd, etc. What I find interesting is that major Linux developers don't copy those features from OS X. I have several theories as to why this is.
- It may be that most of these features are useful for a desktop OS, and most linux developers are focused on Linux as a server and, in fact, will try to block anything not useful for use as a server because for their purpose it just adds bloat and/or instability.
- Linux desktop developers may have never used OS X and are unaware of what these features are or how they work. Many Linux users who do use OS X, seem to become switchers and no longer need to scratch that itch since they get the features just by using OS X.
- Linux desktop developers have a Not Invented Here syndrome and are not willing to even consider technologies from Apple, instead concentrating only on Windows as an alternative they need to keep up with.
- Many of these features are major architectural changes and/or are not backwards compatible and it may be that no one Linux distro has enough momentum that it can make such major moves that might make it less compatible with other versions of Linux and attempts at advocating such standards via groups that span distros is simply too complicated. Linux has no Steve Jobs to make a decision and push it through (which is very good in some cases and very bad in others).
I'd like to see some OS X fanboys who have a clue about the way OS development works; hint: all the majors copy concepts from each other & rarely come up with original features (they mostly come from research projects)Some of those research projects are internal to a given OS vendor. In any case, there is a difference between copying a unproved new technology from a research project and copying a mainstream feature in use from another major OS. The latter takes less effort and is less risky.
Umm, PDF is an open standard, one with many, many implementations. Adobe's reader is not terrible, but its default configuration is (many slow and bloated plugins). Alternately you can use Foxit on Windows, Preview on OS X, XPDF of KPDF on Linux. There are actually quite a few more readers if you want to try others. You can make PDFs using any application on OS X, using CUPs on Linux, MS Office, OpenOffice, Wordperfect, Docbook, etc.
There is a bug in several Linux distros (including Ubuntu) where installing the latest Flash fails silently and tells you you have the most recent when you do not. You need to install by hand. In any case, it does work if you have everything installed properly. I tested it with Flash Player 9.0.115, Firefox 3Beta4, and Ubuntu 8.04.
I didn't say "talk" I said "chat." On Slashdot I should think most people would understand the difference. Reading aloud the characters in a URL you want everyone to visit is a lot less efficient than just sending a link to everyone via IM. Ditto for sending a file to everyone, or really any other local networking.
I agree with this. From what I've seen it seems like there are several internal developer types at Apple:
It seems like at some point Apple needs to step up and form a real security penetration testing team, that will spend all their time trying to hack OS X and Apple products. Heck hire Charlie Miller and pay him to keep on doing this. Give him some stock options and make use of his expertise.
I'd really like for Apple to "step up" and take this bull by the horns and let the world know that they are very serious about security and eliminating *any* means of intrusion, either automated or user driven... and not just rely on the FOSS community to remedy the security problems in the software that they have incorporated into the OS.Actually, Apple has been doing some serious work for advancing security. Leopard has a mandatory access control framework built in, that can be used to sandbox applications. They even sandbox a few services by default adding another layer of security, just as vista does. They do not, however, sandbox Safari by default. Leopard also has an application signing framework, that can be used to help authenticate code origins, an important area for the direction the security industry is taking.
Just as long as they don't implement some Vista like "Allow or Deny?" crap... God that would drive me *nuts*!This security "feature" in Vista is MS's usual user interface design nightmare. They can't seem to wrap their heads around UI design as part of security and seem focused on making sure the user can be blamed instead of just trying to reduce the chances of an exploit working. I would love it if Apple brought restrictive, default MACLs to Leopard, but did the UI and MACL generation properly. You shouldn't be asking users over and over to click Allow or Deny and thereby conditioning them to click Allow to "make stuff work. MACLs should come from Apple and other security vendors. Users should be asked who they trust (Apple, ClamAV, Symantec, Norton, etc. ). Applications should be restricted according to MACLs that ship from the software creator and from security vendors based upon user preferences. Adding security should be easy as buying a subscription to Symantec's MACLs and deciding how much weight to give them. Symantec should install no software... just supply data in an open and defined protocol. Users should almost never be asked to allow or deny a given behavior from an application and only then if it is non-mainstream software and it is trying to do something very unusual - kernel patches, UI hacks, and malware should be about all they are asked about and then the question should be informative with explicit verbs for button, eg, "Program 'Martian Hunter' wants to access your address book (Stop it from accessing your contacts)(allow it to access your contacts once)(allow it to access your contacts always)." Further, whenever possible the OS should try running applications denied access without granting privileges and if that fails run them handing over bogus data and if that too fails try running them in a VM with bogus data and bogus access to other resources before failing entirely. In this way users don't associate granting access or not granting access with whether or not they can use an application. Users want to run stupid little applications and the OS needs to allow them to do so (even if they are malware) but in as safe a fashion as possible.
I agree with your opinion on this. Safari+OS X is lacking several security mechanism in use on Vista+IE by default. Firefox+Ubuntu is likewise lacking one layer. That said, in many cases there are performance and ease of use tradeoffs with implementing additional layers of security. Further, this contest was about targeted attacks by "crackers" which is an uncommon situation for normal consumers.
Because of our malware ecosystem, I'd argue that the appropriate level of security is different for different operating systems and other software and use cases. Normal users don't worry about attacks by a dedicated individual going after just them, but rather about automated worms and widespread malware and phishing. Apple did not lock down Safari in a sandbox. They did not include their anti-phishing feature in a regular release of Safari. I would rather they had done so. But it is not a real concern for most users simply because malware and even phishing is not a major concern to most OS X+Safari users. The risk of a worm or regular phishing attack is very small, so small in fact that to date installing antivirus software that might malfunction has proved the greater danger. While I don't know about the number of phishing attacks, I'd argue that the risk of being compromised is greater if a user switches to Firefox, because there have been in the wild malware attacking Firefox, whereas that is not the case with Safari to date. I'd further argue that it is probable that sandboxing Safari in OS X would have caused more problems for users (due to configuration issues, or malfunctioning sandboxes) than releasing it. Likewise for sandboxing Firefox by default on Ubuntu.
A lot of people complain about Microsoft's security problems with Windows. I think they have a very valid point. A lot of people tout the security features in Windows. They have a very valid point as well. The disconnect and apparent contradiction is explained by looking at the context and how it affects the average user. Windows has not, to date, had an appropriate level of security such that it is not an issue for the average user. Both XP and Vista have suffered from failure to stop "in the wild" automated attacks that caused problems for users. Both have also suffered because misconfigured or malfunctioning security features have caused problems for average users. Microsoft has a lot of work to do in both of these areas.
Both OS X and Ubuntu to date have not had significant problems for normal users due to security failures or malfunctioning security features. This is to their credit, but has more to do with delivering the correct level of security to suit their users (both by default and hardened versions for higher risk deployments). That should not be confused with claims that they have more or better implemented security features. If MS's monopoly market share ever should fall, both those OS's will have a lot of work to do... but of course everyone would also have direct financial interest in making those improvements.
Well my configuration is certainly different. I have Flash Player 9.0.115, Firefox 3Beta4, and Ubuntu 8.04. I had to manually install the Flash player because of a known bug. Apparently Adobe released a new version of Flash 9 in December and on many systems the install checks the MD5 checksum, sees it is different, and silently fails while reporting that it was installed. You might want to try installing Flash by hand or just trying this site again in May, after all the distros manage to patch this. I would argue that since it works for me on Linux, your problem is probably a bug in the version of Firefox or is an issue of an outdated version of Flash combined with a bug in your package management.
This certainly does bring up a long standing complaint I have with every desktop Linux distribution I've tried. Linux is in the lead as far as package managers are concerned... and with ease of finding and installing open source, freeware. At the same time, package managers on Linux are terrible when it comes to commercial software, especially shareware/payware and behind the curve for software portability. I really wish Redhat, Sun, IBM, Canonical, and Apple would all sit down together and create a standard package format based upon GNUStep, and define a set of protocols needed for discovery, transfer, compilation, registration, licensing, installation, and uninstallation of software. I fear, however, that Linux's use as a server and appliance will prevent any such advancement under the auspices of preventing "bloat" with an unhealthy dose of Not Invented Here Syndrome.
The XO was designed and tested to work in conjunction with other XO laptops and with the XO server. Connecting to generic wireless in the states is nice, but you must admit not at all what it was designed for.
This problem certainly took some shine off an otherwise cool little computer for kids.I've heard a lot of stories about people buying XO laptops for their own kid or to play with. I certainly don't have a problem with that, but I do worry that a lot of opinions about the XO will be formed from this type of use, which is not the use for which it was designed. If all the kids in a group don't have one and a server to connect to periodically, well you've just lost a huge chunk of the functionality.
If only that were true. Do you know how much work it is to start a group chat with everyone in say, the coffee shop? Windows auto-discovery of local services/users sucks badly, especially compared to the XO laptop.
And you didn't think upgrading your Flash player would be a good idea? It works fine for me with Firefox and Flash 9 under Ubuntu. (Mind you Flash installation, or really any commercial software, could be a lot easier.)
It's called "humor." You're right, of course. Also we didn't want them to know we were working on such a device for other reasons (sabotage, etc.). I actually used to know a brilliant mathematician who had worked on the project. Sadly he was suffering from Alzheimer's disease and while he could still remember all the trig tables, he occasionally would forget where he was or who he was talking to. Still, he had some great stories. PBS did a good series on the project as well. Requisitioning gold from Fort Knox to use for wiring was my favorite bit of trivia.
As you sow so shall you reap.
After reading the tenth or twentieth scientific article that interviewed people working on the LHC, that includes some wild speculation about remote possibilities that might come to pass when it comes online... this surprises me not at all. I understand being a bit sensationalist to make a more entertaining article. I understand hyping the potential a bit to help keep that government funding coming in. Still, black holes, strangelets, cascading subatomic events, time travelers finding the earliest point to return to... it was a bit much. Maybe you get promoted in experimental physics by making waves and smoking pot with the boss. The you want your name in a magazine so you spin some half-assed idea as though it was a real possibility. The only problem is, some people listened and are now worried.
This is why the Manhattan project was top-secret: two out of six physicists think it might destroy the planet... okay those are good odds, let's try it.
From the summary:
What many of these kids won't get is Internet access: about 90 percent of the villages lack it, and may not get it anytime soon.'"This is a serious drawback as the internet is a great source of information as well as a way to commercialize the computing efforts of these kids and (potentially) give them some good old capitalist reasons to study hard. Even so, while the lack of internet access is not as big of a drawback as it might be. These laptops will presumably still form their own mesh networks and connect to the school's XO server. I bet a lot of Windows using American kids wish their computers would allow them to network with friends nearly as easily.
Best of luck to these teachers. It is always hard being the guinea pig for a new technology and it will probably take a lot of dedication to alter their teaching methods to really take advantage of this new tool.