Imagine my surprise when I try to print a document, and nothing comes out of the printer. The printer isn't claiming it's processing the document - and in one instance, while waiting for my document, I saw the printer fire up and spit out someone else's document. On these occasions, if I go back to my desk and flip back to the Office application that was trying to print, only then does the document print. In fact, I can see Word finally display the status message "Printing page 1 of ..." at the bottom of the window.
Read the rest of the horror story here at Linux in Exile.
My own worst nightmare story with Windows: This was Windows 95, IIRC. The printer driver did two things: 1: It converted a document (or received a document so converted from some other Microsoft app) such that all byte values were being interpreted as control codes (a sort of frame shift, in DNA language). This happens now and then in Windows. This is when the printer spits out several pieces of paper and most of them have one symbol .... like a happy face or a club (as in the card suit) or five or six random sensless blobs ... on each page.
The other thing the printer driver did was to become unattached to the system, but still live. So, it could not be reached, even on the command line. The printer driver survived reboots as well. It could not be killed.
And the document sent to it was a 200 page manuscript.
Solution: Reinstall Windows.
Reason why this is bad: I was printing out a document that needed to be produced some time within the next couple of hours.
Actual solution: Find a different computer, print the document out there, return to original computer, and install Linux on it.
- Log in to post comments
More like this
That's pretty tame by Windows standards as printer problems go. I've seen a Windows machine crash after something went wrong with the printer. I've seen this occur on multiple occasions. The best explanation I've heard is that some Windows drivers keep some of the necessary information to function on the printer. This doesn't seem to be the fault of Windows per se. Just really bad design that somehow only occurs in Windows machines.
wow. I'm not sure what's more sad about this story.
The fact that you're talking about Windows 95 in 2009
Or the fact that you couldn't figure out how to install a printer driver and so changed your OS for one that requires much deeper fiddling to make peripherals work.
Very slow news day i guess.
Anywho, the original article most likely involves someone with pre-Vista printer drivers installed on their Vista machine. Probably the fault of the printer manufacturer not releasing Vista drivers in order to force people to replace their printers ... Not windows fault.
But I'm sure that as a Linux user, you just write your own drivers from scratch now don't you?
Well, you want to talk about printer hassles? How about CUPS? Geez.
Printing with CUPS? There are two steps (if it's a network printer, skip to step 2)
1) Plug the printer in to the computer
2) Print to the printer (Print->choose the printer).
I haven't had much trouble with printing, but what I have noticed (a few years ago) is that linux/bsd/unix systems sometimes require more setup. The trade off is that they don't break if you look at them wrong - they just keep working.
With modern systems, the software can usually figure out how to set itself up AND it works reliably.
Ken, thanks for the helpful comments.
The requirement that Windows be reinstalled in order to turn off the printer driver is what I was told do do by the person on the other end of the support line at some company called Microsoft or something.
Windows fault.
The computer in the post to which I refer did not hvae an old Vista driver sitting there. This is an engineered image at a major institution supported by technical support with windows experts on staff and direct involvement of Microsoft in all the development and training.
Windows fault.
As a Linux user, why would I write my own drivers? There are more drivers for more hardware --- drivers that work, and which, if anything goes wrong with them, are instantly fixed --- than for any other operating system.
Anon ... OK, let's see. I've just installed Linux on this computer I'm typing this comment on. I've got an HP deskjet printer sitting here with a USB cord sticking out of it. Maybe I'll just see what happens if I plug the USB cable into this computer.
Oh, look, a printer Icon has apeared in my thingie-bar up on top, and a brief message identifying the printer correctly tells me that it is ready to go.
Now, I'll try to print the page I'm writing this on.
... file ... print .... printing.
Zero installation effort other than plugging the printer in. Amazing. They should call it something. "Plug and play" maybe.
Anon, a few years ago is a century!
hahaha nice
You fell for the old Windows tech support answer of 'reinstall windows'.
Whenever it gets to that point it means MS tech support doesnt understand the problem or that they don't believe they can get the right information from you to properly identify and fix the problem, so lets do the nuclear attack and wipe everything, start fresh because we know it will eliminate the bad driver at least until you reinstall it.
So you let them change a printer manufacturer's fault into your fault for following bad advice.
Regarding the minor glitch that this story is based on ... You know that 'engineered images' are not perfect, there can be slight hardware differences between the computers the image is meant for. Even minor hardware revisions on supposedly identical pieces can do it. Windows fault ... maybe, but likely you've just been blinded by the titles your IT guys have given themselves.
As for writing your own drivers ... It was a more common thing, say 3 to 5 years ago for sure, you know, back when common Linux users who couldn't dig into the driver code had to shamefully hide how his system lost it's sound card on every reboot, how if you were lucky the CD Burner worked for reading discs but had no support for burning discs, video drivers for any video card that wasn't absolutely mainstream and common handled only the most basic of video functions if they ran at all, and any non-standard peripheral had no support at all. What great days those were huh!
I'm just ranting like this because it's just so laughable that this topic, scrapped from a minor posting on a backwater blog was even somehow newsworthy at scienceblogs.
Am I in love with Windows (vista or other)? No. I guess I'm just the lucky Geologist / CAD / GIS / 3D Mine Modeling guy that hardly ever runs into the Windows issues that Linux (and Mac) people scream about constantly.
Really makes you wonder how many Linux and Mac people just didn't know what the heck they were doing with Windows and took the easy default of blaming the car instead of the loose nut behind the wheel.
"a few years ago is a century!" - said the blogger who had to think back to Windows 95 for his anti-windows anecdote.
Ken said:
I would expect much better from a company that has the largest base of installed systems. If they're that disingenuous, they aren't worth my time or money.
If you had read Greg's introduction to the "Linux in Exile" blog several months ago, perhaps you'd understand.
There are plenty of good reasons to blame Microsoft. Not giving a shit about their end users is the best one.
>blockquote>"a few years ago is a century!" - said the blogger who had to think back to Windows 95 for his anti-windows anecdote.
Give me a freaking break. Troll is seen as trollish.
no Dan, ... an opposing view is seen as trollish apparently.
I've got nothing against Linux, there's many great things and brutal things about Linux just as there is for Windows.
My problem is the people who can't see the fact of the brutal things in Linux. I see you skipped over my chapter on the lack of support for much much hardware just a few years ago.
But I'll let that pass for now and leave you with three words: Open Source Documentation.
Fix this glaring shame ... Then you may attack another OS's help system.
Sorry, Ken, it's been a troll-filled day and I see them everywhere I turn tonight. (Even IRL today.. trollish drivers suck!)
Yes, hardware support was very, very sketchy just a few years ago. Even today, there is a lot of hardware out there that just isn't going to work. How many Linux users have had success stories with Lexmark printers? Not many. HP printers, on the other hand, have always been a breeze for me with Linux. Like you mentioned in your first post, though, I think this is more a problem with the hardware manufacturers than core Linux development. Many manufacturers still won't cooperate with the Linux community to promote open-source drivers (not wishing to let up on their own intellectual property), but don't release their own Linux drivers. They still see Windows as their bread and butter, and no monetary drive from Linux.
Yes, documentation for a lot of open source software is very poor (at best). A lot of it falls on the other side too, being extremely well documented. Developers are usually the last people you want writing documentation for a piece of software, and it's tough to find people who are good writers willing to write docs with no compensation. I'm at a loss on this. I don't see a good solution to the problem other than increased use of Linux requiring better documentation in the future. But without the good documentation, many people won't use it. Catch 22.
I'm still anti-Vista. I use it at my primary employer's location only because it came pre-installed on the system I use, and won't be changed until the warranty is gone. I think it has Alzheimer's (and this is at a retirement home). It seems to forget that some USB devices are connected, but not under any specific conditions. (Could very well be a driver issue from the manuf.) Windows Explorer seems extremely unusable for me. I haven't researched how to get it to display the left pane in the way I'd like, but the default is maddening. So much of it seemed to go backwards as far as general usability is concerned.
I agree with you regarding Windows Explorer in Vista, it also wont lock the view type (details, icons, etc). The folder option to do so doesn't stick.
The USB thing ... I've noticed on network drives Vista doesnt ping the drives constantly, leaves them disconnected until needed, which does cause hiccups for some software occasionally, I suspect it treats USB drives the same way, letting them time out.
It's the same on the Windows side. The fundamental problem is that a lot of hardware manufacturers ship really bad drivers. There's very little an OS can do about badly written third-party software.
I've never had any problems with CUPS printing, it has always auto-detected my HP printer immediately. CUPS is a huge step up from the old days, when you still had to manually set up filters etc.
And just yesterday we wanted to share the printer and set up my girlfriend's laptop with Ubuntu to use it remotely. After activating the sharing of my printer in the CUPS web interface, it popped up within seconds on my girlfriend's laptop as an available printer - no configuration needed at all on her laptop. I was impressed.
The best explanation I've heard is that some Windows drivers keep someseo of the necessary information to function on the printer. This doesn't seem to be the fault of Windows per se. Just really bad design that somehow only occurs in Windows machines.
The big difference in documentation between Windows in general and Linux in general is that Windows' source code is closed, and they hire tech doc writers to produce the documentation that they get. Linux's source is open, and for developers who already know how to read source code, sometimes well commented code is all you need. Also, they depend on free contributions, so documenting Linux isn't going to happen as often as bugfixes or patches or new features because documenting a project isn't a sexy job; it doesn't lead to recognition or add any real value to the end product itself except to people looking to support it.
This is a huge hurdle for Linux to clear, because documentation is necessary for people with middling understanding of computers to be able to make the switch. Total newbies can be plunked in front of an already-configured Linux system for the first time, and get by just as easily as they can in front of Windows for the first time. Experts have a natural desire to tinker and play around with computers and components on the bleeding edge, and occasionally have problems with new hardware that the manufacturer hasn't bothered to make drivers for yet; sometimes, they write their own drivers, then release those drivers to the wild for others to use, and that's how Linux gains support for hardware that's new to the market. It's the middling computer users, who know how to maybe install programs and drivers without help, but who have trouble with actual troubleshooting of Windows issues, that really suffer in the Linux world because a) everything's different from what they're already used to and have probably already crystallized, and b) the resources they're used to using to figure things out are otherwise unavailable. O'Reilly and other publishers of open-source software books sort of fill this gap, but users are used to having the documentation for free and the operating system at cost, rather than the other way around.
My description of the driver situation is not to expect everyone to write their own drivers, but rather it's incumbent on the manufacturer of the hardware to build their drivers in such a manner that either the specifications of the hardware or the source for the drivers themselves are available to Linux developers. But they won't do it if there's no extra money in it, and that means adoption of Linux on a wider scale such that Linux users would prefer to buy their product as a result. This is already seen in the NVidia and Intel graphics card worlds, where the ATI binary blob (while proving ATI cares a bit about Linux users) is usually considered a slap in the face to the openness of the whole operating system's ethos.
As for my own anecdote, I have an HP LaserJet 1020, and after installing Ubuntu, all I had to do was make sure I installed HPIJS from the repository (so it could get the firmware from the website, and upload it to the printer every time it was started -- under Windows, the driver is expected to upload the firmware to the printer every boot, which technically makes it not firmware), and it worked perfectly. This is approximately equivalent to popping in the HP CD provided with the printer, and installing the drivers via the on-screen prompts, only it took a bit of personal knowledge and Google searching, because nowhere did HP actually tell me what package it needed to have installed in order to work under any other OS. Again, this is incumbent on the equipment manufacturer to provide us with the info we need, and says nothing about the OS itself.
If I ever get my hands on a time machine I will visit you and tell you about c:\windows\spool\printers\ and disconnecting the printer.
btw, "web tasarım" is a spammer, you may want to remove his comment.
net stop spooler
delete offending driver
net start spooler
install driver
The spooler, which was still running and trying to print a borked document, had disappeared from any access via the user. It was therefore impossible to stop it. The only way to make it go away would have been to print the thirty or forty thousand pages of paper or to reinstall the system.
Ken: I sense that you have some training from Microsoft, because you really know what you are doing. When it comes to blaming the victim.
Yes, as Stephanie points out, I assure you that the user of the machine in question from LIE knows what he is doing and the people who set up the system do as well, except for the sad fact that someone, somewhere, has been paid off to use Windows only.
Did I say that out loud?
Hank: and what a clever monkey this particular spammer is! This is the first time I've ever seen a spammer auto-quote something that already passed the spam filter, then insert the link into that snippet. Short of searching for intentionally duplicated text and the insertion of a link, I can't think of any way to filter out these goobers without accidentally taking out people who quote others and people who just post a link as their rebuttal.
I assumed as much, just begin a jackass I guess... But in theory, some service / process had to be running the diver, and you could shut it down, if you could figure out what that process was. Although I'm not too sure how easy that would be in Windows 95, XP would be cake.
The core Linux CUPS system, with support for dozens and dozens of printers... 2MB. HPLIP, the HP Linux printer support module from HP... 300KB. Add maybe (and I'm being generous here) 5MB more for various support libraries, and you're up to around 8MB.
HP PSC 3200xi full Windows driver: 300MB (actually, ~800MB unzipped and installed). I kid you not. My jaw dropped when I read the manual. (Basic driver: 31MB.)
Ken wrote:
I'm the Linux in Exile guy - hi there! Actually, I do know what I'm doing on both platforms. Before I became a manager, I was a UNIX systems administrator. I still hold an RHCE, hasn't expired yet. And around the time I was doing UNIX sysadmin, I also doubled as the PC support tech at our office (same office I'm at now) supporting the Windows laptops and PCs. I went through A+ back then, too.
Outside of that, I founded and worked on several open source software projects. I prefer to stay private, so I won't name the projects I created. But you know some of them. I also contributed patches and updates to GNU Emacs, several F/OSS games, and tons of [mostly file] utilities. (Yes, these patches were accepted into the source trees.)
So I'm not complaining about the broken behavior of Windows because I lack any kind of technical knowledge.
You may ask "Why are you blogging about Windows?" Every few months, you'll see a staff writer for some tech magazine claim he's going to try Linux exclusively for a month or so. When the "experiment" is over, the writer usually has lots to say about how this or that thing doesn't work "right" in Linux, because it doesn't work just like Windows does. If mag writers can do this with Windows-Linux, I think it's fair to do the same with Linux-Windows.
I've been fortunate enough to be able to run Linux at work since about 2002. My previous bosses knew about this, and supported it. A few years ago, we got a a new boss who didn't see things exactly that way. So I was asked to move back to Windows, at least for work. The difference between Windows and Linux has been shocking, to say the least. Since I find it interesting when long-time Windows users experiment with Linux for the first time, I thought it might be equally interesting for this long-time Linux user to blog about my first experience running Windows in over 6 or 7 years.
When I blog about something being "broken" in Windows, or something in Windows that is confusing, it's because that feature really is broken or confusing. At least, compared to Linux.
More from Ken:
Nickname_unavailable suggested the same thing in a comment on my blog - that the driver might be for XP, and I'm running Vista.
Here's my reply to NA:
In this case, I know the printer driver was for Vista. When the new printer arrived, our tech support guys hadn't put the driver on the network yet. So they did the next best thing: they put the install CD next to the printer, so those of us who needed to print to that printer could install the driver.
The driver CD helpfully had a separate directory for "Windows XP" and "Windows Vista" (and "Mac OSX", IIRC). I installed the driver for Vista.
So no, this is not a case of pre-Vista printer drivers installed on a Vista machine.
But I have to ask - why would Vista allow itself to install drivers for another OS (Win XP) if those drivers are not compatible, anyway? Vista allowing itself to install known-broken (i.e. non-Vista) drivers doesn't make sense to me. You'd think Microsoft would protect themselves against that, if mixing drivers was such a bad deal.
Vista allowing itself to install known-broken (i.e. non-Vista) drivers doesn't make sense to me. You'd think Microsoft would protect themselves against that, if mixing drivers was such a bad deal.
but...but... but... then, if they did that .... they would not be confusing any more.
And the mystique would die.
Ken: FWIW, the "Better with Windows" campaign isn't a hoax. I saw it over the weekend when I was looking for a hard drive on Newegg. (http://promotions.newegg.com/ASUS/041609/?cm_sp=Subcat_Netbooks-_-ASUS/…) (http://ubuntuforums.org/showthread.php?t=1173213&page=12)
Any better places to shop than NewEgg?
Having just installed a new HP Deskjet under Linux, I have to say it was a breeze. Plug in, and go. Look at Windows/Mac installation CD, and think of uses for it. Coaster or frisbee?
BUT, on the other hand, I never could get my old Lexmark to work under Linux. Still, at peanuts for a printer/copier/scanner, who cares any more?
This doesn't seem to be the fault of Windows per se. Just really bad design that somehow only occurs in Windows machines.