It will be interesting to see how many people take the Linux plunge and break from the swirling vortex of regular, forced product updates. I am betting very few, unfortunately. It's just too much of a leap for most people...when Windows XP/20XX offers such a warm fuzzy UI feeling.
I switched from NT to Samba running on Debian over a year ago. I'm not stuck relying on some company to deliver on-time updates. I've never had a virus infection. Oh, and the only time I need to reboot is to update the kernel (which isn't very often). Talk about a warm fuzzy feeling.
If you just need a cheap box, multiply the cost by two and plan for redundant systems. They will fail sooner or later.
I agree with this. If you build yourself (or get local oem to do it) using commodity parts, you're going to spend a lot less than if you buy a Dell or IBM server. You can then put up two and have true redundancy, and easily obtain parts (or even buy a whole new machine) with no notice.
It's great to have a 24hour replacement policy from Dell or whatever, but that's still 24 hours your mission critical server is down.
For an example, look at google. (apparently they don't even use cases).
Does this actually sync? I remember in rez we setup a shoutcast server in order to play the same music everywhere, but it would always get out of sync, which was pretty annoying.
I've had "single sign-on" for a while now, using Samba as my PDC (originally replaced my NT server about 3 years ago). It wasn't overly difficult to set up, but basically it's running LDAP at the very bottom, and Samba users LDAP as it's database. I can also authenticate from other linux boxes directly against the LDAP server.
I also integrated a number of web applications into it so they authenticate against the LDAP server as well. This isn't always quite as nice - you usually have to type your user/pass in again - but at least it's synchronized with your main account.
As far as end-users are concerned, the result is the same. None of my end-users know any difference between running on this or a Windows server, I don't have any more work to do (things seem to break less than they did with NT.. but I never had stats on this so I can't say for sure) and it's a lot easier to get updates now. And above all, it saves us a lot of money in licencing fees.
Granted, nobody does kernel development in VB nor should they -- that's not the purpose of VB. As a rapid application development language it a great choice.
I personally am not a huge fan of VB, due to I guess two things: it's primarily aimed at MS platforms (I'm getting more into cross-platform stuff, mostly web since it's the fastest way) and I dislike the syntax and style of the language.
I think I share your views on writing actual applications though - personally, I think C (and possibly C++) has no place in a GUI application. I like to think about application/business logic when writing an application, not where in memory a variable gets stored. Don't get me wrong, C has it's uses - writing kernels, drivers, API's and other low-level code - but it's not suited for a high-level application.
I haven't yet found the ideal application for this though. In fact, the majority of my development in the last couple years has been in PHP. This includes a couple of web applications, as well as a whole set of daemons that control industrial equipment. (And to the people that scoff at that concept: It was developed in about a year by one person, is running at many sites right now, and has been doing so for over a year now. By contrast, the http://mat.sourceforge.net/MatPLC project has been around for a few years and has yet to produce anything even close to production-ready. Not doing entirely the same thing, but close. I did look at getting involved btw, but it didn't seem worth it).
Really, it's not like I have any real technical objections to VB (I'm sure I'd get over the syntax thing).. I just have concerns over the portability of code.
and VB.NET support for Mono, but they are always pooh-pooed
I'm don't know too much about this, and I'm sure some of it is just people loving to bash Microsoft, but I'd bet that there are concerns over licencing and patent issues. It would be pretty bad if you spent a lot of time developing an implementation in Mono and then Microsoft came along and threatened to sue - even if you had the money, it would probably be a difficult fight, and not worth it considering you could just do development in another language.
If the linux community is not competing with Microsoft, then why all the constant comparisons to Windows about how linux is superior, which applications run on which platform,
What else are you going to compare it to? Obviously if you're using Linux as a desktop, then you're going to compare it to other desktops.
discussions of how to get more users off of Windows and onto a Linux desktop, etc.
Well, I don't think I touched on this, but of course having more users means more market share, which means more developers being interested, which means more applications being developed, and ultimately, more choice.
"Freedom of choice" is a beautiful thing in linux-land, but for someone on the outside looking in, which choice is the correct one for moving line of business apps to linux en masse (gnome/kde; xorg/xfree86; linux/*bsd/dragonfly)?
Well, that's a problem and it's not a problem. When you're talking about moving a business, then I assume you have an IT department or at least someone serving as an administrator. It's their job to make that decision and then deploy and support it. (Of course, now the administrator is removing choice from the users, so perhaps I'm just being hypocritical.:) )
I think that too many choices only results in unending confusion and a complete reluctance to abandon Windows by those who might otherwise make the switch.
Yes, I'm sure it does. But what would happen if there's no choice? Someone might try out Gnome, not really like the interface, and then decide they hate Linux - and they'd basically be right, since it was the only way to get a GUI. In reality, if you don't like Gnome, you can try KDE. Or one of dozens of other desktop environments/window managers.
Remember too that these are individual developers working on independent projects. If it was Microsoft that had 14 different GUI's to choose from and it was confusing users, then Bill Gates can always step in and stop all but the best GUI from being developed. In the FOSS world, there's no one that can do that.
On a more personal note as a VB.NET developer, I am further put off by the superioristic attitude of many in the *nix community that I should abandon my language of choice and use java/c/c++/python/etc -- and then in the next breath I hear that linux is all about choice.
Well, you're looking at it from the wrong level. Using VB.NET, you're stuck with Microsoft, as are your users. You've removed their choices. If Microsoft decides that they're going to charge a $100,000 per application licence fee, then you're either stuck paying it or you give up development and start from scratch in a new language. If they release Longhorn and active a backdoor in all earlier versions of windows that blocks them from using the internet, your users either have to buy the new Windows, or give up on their investment in your program.
In reality, these are unlikely to happen, and perhaps a bit extreme, but things like this are possible.
All that said, there are a diverse collection of people in the "linux community" -- the (often loud) ones that boo at anything Microsoft are not representative of everyone.
maybe it's simply beyond my understanding why there seem to be dozens and dozens of distros when there is only one linux kernel
Yes, there is only one kernel (that I know of). There are many distros - some have totally different architectures, some are just slightly different ideas, some have different ideals, some are designed for a specific purpose or fill a niche, others are almost identical and have no reason for existing besides that someone made one of them. The real answer here is choice and freedom - people have the freedom to create a distro if they want, and theres lots of choice to use to decide on one.
It provides a strange situation. On one hand, as an experienced user, theres lots of choice. If there's something you don't like about a distro, then likely someone else has already created a distro that fixes it and you can use that one. On the other hand, as a new user with no experience, it's overwhelming. Pretty much you just need to pick a mainstream one (easy to find help that way) and try it.
Also beyond me is the schism between KDE/Gnome, XFree86/X.org, etc. It seems to me that if the Linux community would just bury the hatchet and agree on a best of breed cross section of all the various options in building a system y'all would have the boys in Redmond over a barrel
This is exactly what each distro (in most cases, anyways) thinks they're doing - picking the best tools and putting them all together. Different people have different ideas of what's best, hence the number of distros.
As far as having two big projects working on the same task - ie, Gnome/KDE - there's a lot of issues with that. You can't expect a bunch of developers to suddenly decide to give up on all their code and start working on something else. You also can't just expect them to integrate - in fact I believe they're coded in different languages - the time required to port code would probably be more than just rewriting from scratch.
The big thing here is - no one is competing with Microsoft. There is not some broad community trying to compete with any company or market. It's all individual developers working on projects that are useful to them (maybe sometimes this means financially) and developing tools they need. Often this happens collectively, as many people need the same thing.
Another way to look at it again is having choice. With proprietary software, all the same arguments about architecture and design decisions take place, you just don't see it. They make a decision earlier on, the losing side doesn't actually get to develop much (or any) code, and as a user, you're forced into using the solution the manager liked best (or the solution that had the best presentation). With open source, you get to decide for yourself which is best.
Eventually, one solution will draw more users than the other, and developers will start moving, and eventually you'll get a dead project if enough people switch over. KDE and Gnome are at basically an even split right now, but eventually one will be more predomanant than the other.
Until then, use and/or contribute to the one you like best, and enjoy that you are free to make that choice.
I'm not a pilot, but one thing I didn't like in this article was the talk of "relying" on one. It seems to me, the parachute would be the absolute LAST option - if there's no way you can land the plane on your own, then you pull the lever and hope the parachute saves you.
The article talks about the families suing the company because the parachutes didn't work. It's not like the parachutes killed them.. if there was no parachute, then the plane would still crash.
With RPM, it's easy to add your changes to a package while keeping them easily distinct from any upstream changes.
I have a debian-based system that accomplishes this. I'm not familliar with RPM's way of doing it, but I simply create my own package (with a new name) that depends on the original, and modifies config files during install, or adds new files, or does a dpkg-divert on original files and replaces them with its own (using divert ensures you can install/upgrade the original package and not have it overwrite files you've replaced)
How about just get one that will take a pic every time there is movement, then if someone DID break in, you would have pictures of who it was, plus insurance company can't say you didn't take enough steps.
It would be nice if there was some kind of open-source project that wrote software for this kind of thing, too...
SonyBMG plans to begin the new year with a batch of freshly pressed music cds that will feature copy protection technology from First 4 Internet. First 4 Internet's technology encodes the music files with a heavy encryption that allows standard cd players to playback the music. There are also additional data files on the CD that further enhance encryption. All of which is easily bypassed by simply holding down the shift button when you load the CD into a PC.
Actually, its more like Philips Screw drivers, and Flathead.
Flathead screws are probably the worst invention ever. Philips aren't much better -- mostly because they wear easily.
Some day you americans will see the light and start using robeterson.
That said, Fuller makes an awesome 7-in-1 screwdriver that contains #1 and #2 robertson, philips, flathead, and 1/4, 3/8, and something else driver bits. All on one shaft that is reversable, with reversable bits on each end. I almost* always have it in my toolbelt, and it's probably the most useful tool I've ever owned:p
* almost always, because a couple of days ago someone in the office stole mine. When I find out who it was I will be introducing their fingers to a pair of vice grips.
The real question should be that if we could get the windows kernel and only the windows kernel, how many bugs in comparison would that have?
The point would still be made. They're doing an average, so roughly you'd expect that the percentage of bugs is about the same even if you were to isolate pieces of code. Now, many people have pointed out that if the 'extraneous crap' is running in kernel space, then it should be counted as the kernel (such as the GUI, IE, etc) which I think is fair.
But anyways, let's assume their actual "kernel" is only 10 million lines of code. If it has a lower percentage of bugs, that means the rest of their proprietary code has an even higher percentage of bugs. Perhaps the kernel itself is higher, which makes things even worse.
Now show me the great Open Source product being able to replace SAP, PeopleSoft and the like and everything will be jolly good.
Oh, I forgot, I'm supposed to write it myself. My stupid.
Attitudes like this are exactly why no open source version exists. There is quite a bit of value in writing an open source solution, espessially if your primary business is not software development. Hiring a couple of in-house programmers for a year to do development is probably not much more expensive (perhaps even cheaper) than paying the proprietary software to begin with, espessailly once you get customizations done.
There's likely enough demand (espessially now that PeopleSoft is going to disappear) that provided your developers know how to do OS (as in, provide all the necessary tools - cvs, mailing lists, etc) and are properly organized (proper roadmap, enough documentation along the way), other companies and/or individuals will likely take an interest and help out with development.
The benefits? You get the software you want exactly, customized to your business. You get enhancements as other companies customize it to their needs and add new features, and once you're done writing it, you don't even really need to continue development if you don't want to, as hopefully there's a community working on it. Likely the developers will leave (or ge laid off..) and go start their own company doing development on the software for other companies. If you ever need to hire developers, there's a pool of them to choose from that are already familliar with the code. The key thing is you never have to be worried about what happens to the company behind the software, because there isn't one.
Of course, you may not want to venture into the seedy world of open source development where rivals can look at the code you're using and totally undermine your business pratices (I hope you can detect the sarcasm). Perhaps the best thing to do would be to go pay another huge sum of money to Oracle, spend a year re-implementing your setup, and perhaps go through this mess again in a few years when Oracle decides to discontinue their software, gets bought out by a competitor, changes their licensing scheme, take the product in a new direction, gets crushed by Microsoft, goes bankrupt, gets sued for stealing code....
Someone wanted to sell his new CRT with a larger "inch number", and so he measured the size of the CRT across its mounting points. Et voila: 2 more inches! TFTs don't have that large mounting points because they are not as heavy as a CRT, so this trick does not work well.
CRTs have always been measured by the size of the picture tube. However, you can't project right to the edge of the tube, so they build the case over top of the unusable edges, and end up with a smaller viewable area (a stat most manufacturers have, although much harder to find). A 17" CRT often has closer to 15.5" viewable area. TFTs on the other hand are 100% usable. A 15" TFT has 15" viewable area.
It's still a dirty trick, but there is at least some merit to the measurement.
I used to do what you do but I had to abandon that idea. That defensive tactic won't get rid of those who setup the spam server to autogenerate millions of addresses to your domain.
I use a subdomain, but otherwise do the same thing. It works well, because the sub-domain doesn't get directory harvest attacks, only the main domain (and I only have a couple valid addresses there). Certainly doesn't keep me spam free, but helps to filter out a lot of it.
When exactly did this whole business of customers getting sued for using a product someone else wrote start? It seems to me this never happened before SCO threatened Linux users. I thought the reason for that was simply with Linux, there is no entity that creates it that can be sued, so they were going after the customers.
Traditionally, it's the responsibility of the software developer to deal with the IP stuff, and the customers can just go and use it. If the developer violates some patents, they're the ones responsible for it. Haven't customers ALWAYS been indemnified? Frankly, when all these companies cam out and specifically said they were indemnifing customers, I thought it was just a PR scam.
As far as I know, the only two cases brought against end users that have anything to do with IP violations in software are SCO vs DialmerChrylser, and SCO vs AutoZone.. both of those basically got laughed out of court and are all but over. Has there ever been a case of a company going after end-users of a software product (instead of the developer) for patent violations?
All too many people think that if you have the right "digital tools" amazing Pixar quality art will pop-out.
I think another problem a lot of programmers suffer is that they think because they know how to use all the little features of photoshop and can make fancy looking lightning bolts and gradient backgrounds, they're graphic artists. What they really end up with is an image that contains all sorts of neat elements, but all piled together to make one big photoshop effects ad that overall screams amateur.
I'm mixed. I've been involved with servers (as an admin) running Plesk for a few years now. To sum it up, I hate it.
They used to be good at listening to requests and bug reports, now they don't. Their QA also seems to be lacking, or even non-existant - we've been trying to upgrade/migrate to a new server running Plesk 7 for a while now. We've actually spent several hours at it twice now, both times when plesk says "oh, we're sure it will all work properly now" and came up with a huge list of bugs both times (and ended up aborting the upgrade, I might add).
One suggestion they gave us at one point was to upgrade the existing server to plesk 7 first, then migrate. We didn't even consider that - besides not liking the idea of upgrading a production server, I didn't trust that their upgrade process even worked at all. Wasn't really willing to bet several people's domains on it, to say the least..
Anyway, one of the other things that really bugs me is the fact it's proprietary. I can't change the code. I can't add buttons to add features specific to our server. I can't fix their bandwidth reporting (ie, how it says some clients use 200k of mail traffic when really they're using 2GB). I can't make a custom DNS zone (that's not even in Plesk) without it getting randomly overwritten every 1 week or 3 months or somewhere in there, or even figure out why it does (did? hasn't happened in a while.. knock on wood) that.
Even if they did provide source, who am I fixing them for? I pay swsoft, then develop for free, and let them sell my changes?
One of the other things I'll point out is that Plesk was orginially linux and freebsd. As far as I can tell, Plesk for windows is a different codebase, with maybe a similar gui - on *nix, it uses Apache, on windows, IIS. I don't run windows servers on the internet to begin with, but I certainly wouldn't consider running Plesk on them, considering the QA experiences I've had with them.
Of course, these are just my opinions, and while I do come down hard on Plesk, I do really like their interface. Strictly as a webhosting client (that doesn't have to do anything out of the norm), it's a nice management interface. Also saves dealing with simple client requests ("can you add an email alias for me?" "oh sorry, that also needs to go to this person..").
Re:Offtopic, but there's nowhere else to post this
on
Press freedom
·
· Score: 1
After reading your journal and following This link, I have found that TheIndividual has managed to remove the binary tags. (If you were already aware of this, I apoligise...)
I found TheIndividual's page when I was first looking for the (GPL'd) source code without having to subscribe to their forums.. Just read through that page and got even more insight into the situation. It left me asking one big question:
Why has no one taken Sveasoft's firmware and forked it? Say thanks for your work, but you are a bastard, we don't agree with the way you're taking the project, so here's our own. Thats how ever other OS project works (well, maybe without the bastard part).
I suppose by posting this I'm doing the same thing I'm complaining about - namely, complaining without taking action. There are a few things that the sveasoft firmware can't do that I'd like to be able to do. Once I get time, I will definately start working on it if no one else has.. but surely I'm not the only one with no time to take on an additional project?
They should just go non-tld and be done with it. That way, companies and orgs won't have to register a half dozen sites just to redirect them to one.
It's nice to be able to print "mybusiness.com" on something and have people know it's a website. "http://www.mybusiness.com" CAN look ok, but for a lot of things, design-wise, it's nicer to drop the 'technical' stuff.
It's also easier to tell people things.. the "dot com" tells them it's a website. As an example, "Look us up, mybusiness dot com" vs "Look us up, AOL keyword mybusiness". (or "web address")
Except latency sucks, and DirecWay's Fair Access Policy means that if you download more than 169 MB in some period of time (their website doesn't say), you get dropped to 47 kbps for 8 - 12 hours.
We're moving our office just out of town, and I'm kind of stuck for internet. Originally I said no to satellite, because I use SSH a lot and the latency would kill me. I would also definately only go on a plan where I could download, ie, a linux iso, with no problems.
Cable company ways $3k to install (which spread out over a year works to paying $350/mo for their crappy service, as opposed to $99); DSL is not available; fiber (while cheaper to install than cable!) is too expensive per month (even if I move my colocated servers in-house); T1's are even more expensive than fiber (and yet, slower); and though a wireless provider is coming 'soon', they're not setup yet.
So I called a satellite provider just out of due dilligance. First thing he asked was "dsl or cable isn't available?". He also ended up telling me that ssh would be faster over dialup..
What we actually decided to do is go on dialup (I think I'm going to be working from home a lot..) and wait out the wireless guys. I know the owner through a friend that used to work there, and I've been talking to them.. right now they're waiting for the township to approve an antenna location, and we have some contacts with the township so we've been telling them that it's quite important to us to finish this asap.
Ergo, as long as the world keeps getting more byte-bloated, the technical limitations of satellite internet will plague it into highly rural markets, as even many small communities have some form of land-based broadband.
Provided you had a large enough subscriber base, and weren't trying to make money, it'd probably be cheaper and faster to bring in a T1 or fiber line and do wireless than for people in a rural community to go on satellite.
Slashdot Mirror
It will be interesting to see how many people take the Linux plunge and break from the swirling vortex of regular, forced product updates. I am betting very few, unfortunately. It's just too much of a leap for most people...when Windows XP/20XX offers such a warm fuzzy UI feeling.
I switched from NT to Samba running on Debian over a year ago. I'm not stuck relying on some company to deliver on-time updates. I've never had a virus infection. Oh, and the only time I need to reboot is to update the kernel (which isn't very often). Talk about a warm fuzzy feeling.
If you just need a cheap box, multiply the cost by two and plan for redundant systems. They will fail sooner or later.
I agree with this. If you build yourself (or get local oem to do it) using commodity parts, you're going to spend a lot less than if you buy a Dell or IBM server. You can then put up two and have true redundancy, and easily obtain parts (or even buy a whole new machine) with no notice.
It's great to have a 24hour replacement policy from Dell or whatever, but that's still 24 hours your mission critical server is down.
For an example, look at google. (apparently they don't even use cases).
sync'ing of songs throughout the house
Does this actually sync? I remember in rez we setup a shoutcast server in order to play the same music everywhere, but it would always get out of sync, which was pretty annoying.
I've had "single sign-on" for a while now, using Samba as my PDC (originally replaced my NT server about 3 years ago). It wasn't overly difficult to set up, but basically it's running LDAP at the very bottom, and Samba users LDAP as it's database. I can also authenticate from other linux boxes directly against the LDAP server.
.. but I never had stats on this so I can't say for sure) and it's a lot easier to get updates now. And above all, it saves us a lot of money in licencing fees.
I also integrated a number of web applications into it so they authenticate against the LDAP server as well. This isn't always quite as nice - you usually have to type your user/pass in again - but at least it's synchronized with your main account.
As far as end-users are concerned, the result is the same. None of my end-users know any difference between running on this or a Windows server, I don't have any more work to do (things seem to break less than they did with NT
Granted, nobody does kernel development in VB nor should they -- that's not the purpose of VB. As a rapid application development language it a great choice.
.. I just have concerns over the portability of code.
I personally am not a huge fan of VB, due to I guess two things: it's primarily aimed at MS platforms (I'm getting more into cross-platform stuff, mostly web since it's the fastest way) and I dislike the syntax and style of the language.
I think I share your views on writing actual applications though - personally, I think C (and possibly C++) has no place in a GUI application. I like to think about application/business logic when writing an application, not where in memory a variable gets stored. Don't get me wrong, C has it's uses - writing kernels, drivers, API's and other low-level code - but it's not suited for a high-level application.
I haven't yet found the ideal application for this though. In fact, the majority of my development in the last couple years has been in PHP. This includes a couple of web applications, as well as a whole set of daemons that control industrial equipment. (And to the people that scoff at that concept: It was developed in about a year by one person, is running at many sites right now, and has been doing so for over a year now. By contrast, the http://mat.sourceforge.net/MatPLC project has been around for a few years and has yet to produce anything even close to production-ready. Not doing entirely the same thing, but close. I did look at getting involved btw, but it didn't seem worth it).
Really, it's not like I have any real technical objections to VB (I'm sure I'd get over the syntax thing)
and VB.NET support for Mono, but they are always pooh-pooed
I'm don't know too much about this, and I'm sure some of it is just people loving to bash Microsoft, but I'd bet that there are concerns over licencing and patent issues. It would be pretty bad if you spent a lot of time developing an implementation in Mono and then Microsoft came along and threatened to sue - even if you had the money, it would probably be a difficult fight, and not worth it considering you could just do development in another language.
If the linux community is not competing with Microsoft, then why all the constant comparisons to Windows about how linux is superior, which applications run on which platform,
:) )
What else are you going to compare it to? Obviously if you're using Linux as a desktop, then you're going to compare it to other desktops.
discussions of how to get more users off of Windows and onto a Linux desktop, etc.
Well, I don't think I touched on this, but of course having more users means more market share, which means more developers being interested, which means more applications being developed, and ultimately, more choice.
"Freedom of choice" is a beautiful thing in linux-land, but for someone on the outside looking in, which choice is the correct one for moving line of business apps to linux en masse (gnome/kde; xorg/xfree86; linux/*bsd/dragonfly)?
Well, that's a problem and it's not a problem. When you're talking about moving a business, then I assume you have an IT department or at least someone serving as an administrator. It's their job to make that decision and then deploy and support it. (Of course, now the administrator is removing choice from the users, so perhaps I'm just being hypocritical.
I think that too many choices only results in unending confusion and a complete reluctance to abandon Windows by those who might otherwise make the switch.
Yes, I'm sure it does. But what would happen if there's no choice? Someone might try out Gnome, not really like the interface, and then decide they hate Linux - and they'd basically be right, since it was the only way to get a GUI. In reality, if you don't like Gnome, you can try KDE. Or one of dozens of other desktop environments/window managers.
Remember too that these are individual developers working on independent projects. If it was Microsoft that had 14 different GUI's to choose from and it was confusing users, then Bill Gates can always step in and stop all but the best GUI from being developed. In the FOSS world, there's no one that can do that.
On a more personal note as a VB.NET developer, I am further put off by the superioristic attitude of many in the *nix community that I should abandon my language of choice and use java/c/c++/python/etc -- and then in the next breath I hear that linux is all about choice.
Well, you're looking at it from the wrong level. Using VB.NET, you're stuck with Microsoft, as are your users. You've removed their choices. If Microsoft decides that they're going to charge a $100,000 per application licence fee, then you're either stuck paying it or you give up development and start from scratch in a new language. If they release Longhorn and active a backdoor in all earlier versions of windows that blocks them from using the internet, your users either have to buy the new Windows, or give up on their investment in your program.
In reality, these are unlikely to happen, and perhaps a bit extreme, but things like this are possible.
All that said, there are a diverse collection of people in the "linux community" -- the (often loud) ones that boo at anything Microsoft are not representative of everyone.
maybe it's simply beyond my understanding why there seem to be dozens and dozens of distros when there is only one linux kernel
Yes, there is only one kernel (that I know of). There are many distros - some have totally different architectures, some are just slightly different ideas, some have different ideals, some are designed for a specific purpose or fill a niche, others are almost identical and have no reason for existing besides that someone made one of them. The real answer here is choice and freedom - people have the freedom to create a distro if they want, and theres lots of choice to use to decide on one.
It provides a strange situation. On one hand, as an experienced user, theres lots of choice. If there's something you don't like about a distro, then likely someone else has already created a distro that fixes it and you can use that one. On the other hand, as a new user with no experience, it's overwhelming. Pretty much you just need to pick a mainstream one (easy to find help that way) and try it.
Also beyond me is the schism between KDE/Gnome, XFree86/X.org, etc. It seems to me that if the Linux community would just bury the hatchet and agree on a best of breed cross section of all the various options in building a system y'all would have the boys in Redmond over a barrel
This is exactly what each distro (in most cases, anyways) thinks they're doing - picking the best tools and putting them all together. Different people have different ideas of what's best, hence the number of distros.
As far as having two big projects working on the same task - ie, Gnome/KDE - there's a lot of issues with that. You can't expect a bunch of developers to suddenly decide to give up on all their code and start working on something else. You also can't just expect them to integrate - in fact I believe they're coded in different languages - the time required to port code would probably be more than just rewriting from scratch.
The big thing here is - no one is competing with Microsoft. There is not some broad community trying to compete with any company or market. It's all individual developers working on projects that are useful to them (maybe sometimes this means financially) and developing tools they need. Often this happens collectively, as many people need the same thing.
Another way to look at it again is having choice. With proprietary software, all the same arguments about architecture and design decisions take place, you just don't see it. They make a decision earlier on, the losing side doesn't actually get to develop much (or any) code, and as a user, you're forced into using the solution the manager liked best (or the solution that had the best presentation). With open source, you get to decide for yourself which is best.
Eventually, one solution will draw more users than the other, and developers will start moving, and eventually you'll get a dead project if enough people switch over. KDE and Gnome are at basically an even split right now, but eventually one will be more predomanant than the other.
Until then, use and/or contribute to the one you like best, and enjoy that you are free to make that choice.
I'm not a pilot, but one thing I didn't like in this article was the talk of "relying" on one. It seems to me, the parachute would be the absolute LAST option - if there's no way you can land the plane on your own, then you pull the lever and hope the parachute saves you.
.. if there was no parachute, then the plane would still crash.
The article talks about the families suing the company because the parachutes didn't work. It's not like the parachutes killed them
With RPM, it's easy to add your changes to a package while keeping them easily distinct from any upstream changes.
I have a debian-based system that accomplishes this. I'm not familliar with RPM's way of doing it, but I simply create my own package (with a new name) that depends on the original, and modifies config files during install, or adds new files, or does a dpkg-divert on original files and replaces them with its own (using divert ensures you can install/upgrade the original package and not have it overwrite files you've replaced)
How about just get one that will take a pic every time there is movement, then if someone DID break in, you would have pictures of who it was, plus insurance company can't say you didn't take enough steps.
It would be nice if there was some kind of open-source project that wrote software for this kind of thing, too...
a bit OT, but does anyone know of tools for admins of shared servers to scan for vulnerabilites in customer-installed web applications like these?
I just went through by hand, and 8 of 9 installed copies of phpBB on my server were vulnerable.
As long as you remove some keys from your keyboard, it's all good.
SonyBMG plans to begin the new year with a batch of freshly pressed music cds that will feature copy protection technology from First 4 Internet. First 4 Internet's technology encodes the music files with a heavy encryption that allows standard cd players to playback the music. There are also additional data files on the CD that further enhance encryption. All of which is easily bypassed by simply holding down the shift button when you load the CD into a PC.
Actually, its more like Philips Screw drivers, and Flathead.
:p
Flathead screws are probably the worst invention ever. Philips aren't much better -- mostly because they wear easily.
Some day you americans will see the light and start using robeterson.
That said, Fuller makes an awesome 7-in-1 screwdriver that contains #1 and #2 robertson, philips, flathead, and 1/4, 3/8, and something else driver bits. All on one shaft that is reversable, with reversable bits on each end. I almost* always have it in my toolbelt, and it's probably the most useful tool I've ever owned
* almost always, because a couple of days ago someone in the office stole mine. When I find out who it was I will be introducing their fingers to a pair of vice grips.
The real question should be that if we could get the windows kernel and only the windows kernel, how many bugs in comparison would that have?
The point would still be made. They're doing an average, so roughly you'd expect that the percentage of bugs is about the same even if you were to isolate pieces of code. Now, many people have pointed out that if the 'extraneous crap' is running in kernel space, then it should be counted as the kernel (such as the GUI, IE, etc) which I think is fair.
But anyways, let's assume their actual "kernel" is only 10 million lines of code. If it has a lower percentage of bugs, that means the rest of their proprietary code has an even higher percentage of bugs. Perhaps the kernel itself is higher, which makes things even worse.
Now show me the great Open Source product being able to replace SAP, PeopleSoft and the like and everything will be jolly good.
Oh, I forgot, I'm supposed to write it myself. My stupid.
Attitudes like this are exactly why no open source version exists. There is quite a bit of value in writing an open source solution, espessially if your primary business is not software development. Hiring a couple of in-house programmers for a year to do development is probably not much more expensive (perhaps even cheaper) than paying the proprietary software to begin with, espessailly once you get customizations done.
There's likely enough demand (espessially now that PeopleSoft is going to disappear) that provided your developers know how to do OS (as in, provide all the necessary tools - cvs, mailing lists, etc) and are properly organized (proper roadmap, enough documentation along the way), other companies and/or individuals will likely take an interest and help out with development.
The benefits? You get the software you want exactly, customized to your business. You get enhancements as other companies customize it to their needs and add new features, and once you're done writing it, you don't even really need to continue development if you don't want to, as hopefully there's a community working on it. Likely the developers will leave (or ge laid off..) and go start their own company doing development on the software for other companies. If you ever need to hire developers, there's a pool of them to choose from that are already familliar with the code. The key thing is you never have to be worried about what happens to the company behind the software, because there isn't one.
Of course, you may not want to venture into the seedy world of open source development where rivals can look at the code you're using and totally undermine your business pratices (I hope you can detect the sarcasm). Perhaps the best thing to do would be to go pay another huge sum of money to Oracle, spend a year re-implementing your setup, and perhaps go through this mess again in a few years when Oracle decides to discontinue their software, gets bought out by a competitor, changes their licensing scheme, take the product in a new direction, gets crushed by Microsoft, goes bankrupt, gets sued for stealing code....
Someone wanted to sell his new CRT with a larger "inch number", and so he measured the size of the CRT across its mounting points. Et voila: 2 more inches! TFTs don't have that large mounting points because they are not as heavy as a CRT, so this trick does not work well.
CRTs have always been measured by the size of the picture tube. However, you can't project right to the edge of the tube, so they build the case over top of the unusable edges, and end up with a smaller viewable area (a stat most manufacturers have, although much harder to find). A 17" CRT often has closer to 15.5" viewable area. TFTs on the other hand are 100% usable. A 15" TFT has 15" viewable area.
It's still a dirty trick, but there is at least some merit to the measurement.
I used to do what you do but I had to abandon that idea. That defensive tactic won't get rid of those who setup the spam server to autogenerate millions of addresses to your domain.
I use a subdomain, but otherwise do the same thing. It works well, because the sub-domain doesn't get directory harvest attacks, only the main domain (and I only have a couple valid addresses there). Certainly doesn't keep me spam free, but helps to filter out a lot of it.
When exactly did this whole business of customers getting sued for using a product someone else wrote start? It seems to me this never happened before SCO threatened Linux users. I thought the reason for that was simply with Linux, there is no entity that creates it that can be sued, so they were going after the customers.
Traditionally, it's the responsibility of the software developer to deal with the IP stuff, and the customers can just go and use it. If the developer violates some patents, they're the ones responsible for it. Haven't customers ALWAYS been indemnified? Frankly, when all these companies cam out and specifically said they were indemnifing customers, I thought it was just a PR scam.
As far as I know, the only two cases brought against end users that have anything to do with IP violations in software are SCO vs DialmerChrylser, and SCO vs AutoZone.. both of those basically got laughed out of court and are all but over. Has there ever been a case of a company going after end-users of a software product (instead of the developer) for patent violations?
All too many people think that if you have the right "digital tools" amazing Pixar quality art will pop-out.
I think another problem a lot of programmers suffer is that they think because they know how to use all the little features of photoshop and can make fancy looking lightning bolts and gradient backgrounds, they're graphic artists. What they really end up with is an image that contains all sorts of neat elements, but all piled together to make one big photoshop effects ad that overall screams amateur.
I'm mixed. I've been involved with servers (as an admin) running Plesk for a few years now. To sum it up, I hate it.
They used to be good at listening to requests and bug reports, now they don't. Their QA also seems to be lacking, or even non-existant - we've been trying to upgrade/migrate to a new server running Plesk 7 for a while now. We've actually spent several hours at it twice now, both times when plesk says "oh, we're sure it will all work properly now" and came up with a huge list of bugs both times (and ended up aborting the upgrade, I might add).
One suggestion they gave us at one point was to upgrade the existing server to plesk 7 first, then migrate. We didn't even consider that - besides not liking the idea of upgrading a production server, I didn't trust that their upgrade process even worked at all. Wasn't really willing to bet several people's domains on it, to say the least..
Anyway, one of the other things that really bugs me is the fact it's proprietary. I can't change the code. I can't add buttons to add features specific to our server. I can't fix their bandwidth reporting (ie, how it says some clients use 200k of mail traffic when really they're using 2GB). I can't make a custom DNS zone (that's not even in Plesk) without it getting randomly overwritten every 1 week or 3 months or somewhere in there, or even figure out why it does (did? hasn't happened in a while.. knock on wood) that.
Even if they did provide source, who am I fixing them for? I pay swsoft, then develop for free, and let them sell my changes?
One of the other things I'll point out is that Plesk was orginially linux and freebsd. As far as I can tell, Plesk for windows is a different codebase, with maybe a similar gui - on *nix, it uses Apache, on windows, IIS. I don't run windows servers on the internet to begin with, but I certainly wouldn't consider running Plesk on them, considering the QA experiences I've had with them.
Of course, these are just my opinions, and while I do come down hard on Plesk, I do really like their interface. Strictly as a webhosting client (that doesn't have to do anything out of the norm), it's a nice management interface. Also saves dealing with simple client requests ("can you add an email alias for me?" "oh sorry, that also needs to go to this person..").
After reading your journal and following This link, I have found that TheIndividual has managed to remove the binary tags. (If you were already aware of this, I apoligise...)
I found TheIndividual's page when I was first looking for the (GPL'd) source code without having to subscribe to their forums.. Just read through that page and got even more insight into the situation. It left me asking one big question:
Why has no one taken Sveasoft's firmware and forked it? Say thanks for your work, but you are a bastard, we don't agree with the way you're taking the project, so here's our own. Thats how ever other OS project works (well, maybe without the bastard part).
I suppose by posting this I'm doing the same thing I'm complaining about - namely, complaining without taking action. There are a few things that the sveasoft firmware can't do that I'd like to be able to do. Once I get time, I will definately start working on it if no one else has.. but surely I'm not the only one with no time to take on an additional project?
They should just go non-tld and be done with it. That way, companies and orgs won't have to register a half dozen sites just to redirect them to one.
It's nice to be able to print "mybusiness.com" on something and have people know it's a website. "http://www.mybusiness.com" CAN look ok, but for a lot of things, design-wise, it's nicer to drop the 'technical' stuff.
It's also easier to tell people things.. the "dot com" tells them it's a website. As an example, "Look us up, mybusiness dot com" vs "Look us up, AOL keyword mybusiness". (or "web address")
Maybe we should propose .crap
Except latency sucks, and DirecWay's Fair Access Policy means that if you download more than 169 MB in some period of time (their website doesn't say), you get dropped to 47 kbps for 8 - 12 hours.
.. right now they're waiting for the township to approve an antenna location, and we have some contacts with the township so we've been telling them that it's quite important to us to finish this asap.
We're moving our office just out of town, and I'm kind of stuck for internet. Originally I said no to satellite, because I use SSH a lot and the latency would kill me. I would also definately only go on a plan where I could download, ie, a linux iso, with no problems.
Cable company ways $3k to install (which spread out over a year works to paying $350/mo for their crappy service, as opposed to $99); DSL is not available; fiber (while cheaper to install than cable!) is too expensive per month (even if I move my colocated servers in-house); T1's are even more expensive than fiber (and yet, slower); and though a wireless provider is coming 'soon', they're not setup yet.
So I called a satellite provider just out of due dilligance. First thing he asked was "dsl or cable isn't available?". He also ended up telling me that ssh would be faster over dialup..
What we actually decided to do is go on dialup (I think I'm going to be working from home a lot..) and wait out the wireless guys. I know the owner through a friend that used to work there, and I've been talking to them
Ergo, as long as the world keeps getting more byte-bloated, the technical limitations of satellite internet will plague it into highly rural markets, as even many small communities have some form of land-based broadband.
Provided you had a large enough subscriber base, and weren't trying to make money, it'd probably be cheaper and faster to bring in a T1 or fiber line and do wireless than for people in a rural community to go on satellite.