Hacker News new | past | comments | ask | show | jobs | submit login
CUPS 2.0 (cups.org)
158 points by tomkwok on Oct 2, 2014 | hide | past | favorite | 60 comments



I wish printing was harder. I have piles of slides filling my desk. Slides on paper, printed by someone else, pages that have been barely read once, their information is already outdated and the contents could have been collectively read from a white screen in the meetings.

Using Linux since 1995 forced me into paperless thinking, because printing used to be hard. And these days, with all those smartphone, tablet, laptop, ereader etc. screen it should be virtually unnecessary in office environments. But still people just keep on printing everything.

I hope there is some terrible fault in CUPS that makes people to think, that trying to print this is not worth the trouble</semisarcasm>


Funny that I have the exactly opposite feeling. I want to print more, but I don't, because CUPS is a kludge, and I don't feel like using Windows at work for the sole purpose of somewhat sane printer handling.

Now to answer why and what I would want to print more? Surprisingly, I mostly want to print code. The reason is that computers suck for reading and thinking, and in my specific case, I can't focus on thinking hard in front of a computer screen (it's too easy to turn on HN).

When I'm stuck with a hard refactor to do, I like to print out all relevant files, take coloured pens and highlighters and keep drawing and annotating code until everything is clear in my hand. Paper is simply superior in annotating capabilities and high-level view (i.e. spread 20 pages in front of you to see everything). I haven't found any tool yet that would make doing this on computer even comparable.

I once asked a very known HN user, who prints out all relevant code every afternoon as a method of planning work for the next day, whether he worries about the amount of paper he uses. The answer I got: "Paper is cheap. Focus is golden."

As for ecology side of printing, I used to worry much more about it before. But at some point I realized that I didn't do any math, and from what I can tell, "paperless world" seems to be much more damaging to environment than cutting down trees for paper, and as long as we recycle, paper ain't that bad. But still, I didn't do the math.


>When I'm stuck with a hard refactor to do, I like to print out all relevant files, take coloured pens and highlighters and keep drawing and annotating code until everything is clear in my hand. Paper is simply superior in annotating capabilities and high-level view (i.e. spread 20 pages in front of you to see everything). I haven't found any tool yet that would make doing this on computer even comparable.

I can't agree more. I tried several ways to get a better overview of projects I work on, but ultimately it's all very klutzy and awkward. When you have 30 (or more) files with cross references and you need to get a global understanding of the whole architecture, working with code can be very hard. Even with visual tools like UML diagram or similar IDE features, even when ctrl+clicking offered by various editors to jump to definitions and jump around code, even with ctags and multiple buffers open in vim. I tried working with 4 screens but my eyes have to move all over the place to keep following, I can't easily re-organize my files next to each other to keep relevant portions of code close together (regardless of logical modularity in my implementation).

Paper is really the best we have at the moment, maybe a more modern environment in the future will solve this (I recall LightTable promising something similar where you could move around small blocks of functions, during their Kickstarter, however that was never delivered unfortunately), but for the moment we are stuck with paper.


I think this is sort of the fact that IDEs are signed primarily for writing code, not for understanding existing code.

If you look at a reverse engineering tool like IDA Pro, despite working with lower level code, you get powerful understanding tools, excellent cross referencing and diagram generation tools, graph views of code so you can see every decision made without having to read them all out and now even stuff like "proximity" view, which gives a nice way to understand calling structures.

http://www.hexblog.com/?p=468


Isn't it better to just draw a diagram and write by hand though? The process of mentally filtering out what's irrelevant to the problem is important as well I think. By the way this is one reason why I like big cubicles as programmer workspaces so much, you have lots of space to hang up your drawings to the sides of your screen.


I haven't had problems printing on Linux in years. I can't remember the last time I had a hard time printing on Linux. I hate printers, because they all suck (well, I had some nice high end laser printers in the past that were great), but In fact, I generally find myself booting into Linux to print and scan because my printer and scanner are old and don't have modern Windows drivers (this is true of several devices, including some that I don't consider outdated...high end 24 bit audio interfaces, USB MIDI controllers, etc.).


> I hate printers, because they all suck

Yeah. This is a topic for a completely separate thread, but I have to say this: I FUCKING HATE printers. From jamming hardware, through broken driver issues to all the shenanigans printer companies do (especially with inkjets).


In a just world we could sue inkjet printer manufacturers for selling something that is so clearly not fit for its intended purpose.


Out of curiosity, how does CUPS prevent you from printing out code? I can understand that there might be some advanced printing features that don't work perfectly with Linux because of lack of driver support. It's been ages since I last had difficulty with printing text. I've had difficulty with old printers and new Windows versions and vice versa.


It prevents me from printing, period. It's better than it used to be years ago, but still there's ton of work to do every time I want to plug into a new printer, and half of the times it doesn't work. I know that it is probably due to my lack of knowledge and experience, but I'm not frustrated enough to sit down and figure CUPS out for good.


Don't worry, not only CUPS is PITA. Even one unnamed OS with totally different print service randomly decides that your printer is offline (what? it is live on the network, I can ping it! I can restart it million times and the spooler still tells me, that it is offline.), just to annoy you. The more important that printed document is and the less time there is for you to present it, the higher probability, that you will not be able to print it.


I see. Most of the time I'm lucky enough to work with HP printers. The hp-setup tool makes installation of drivers trivial. Once in a while I work with Brother or some other brand of printers, and that can require a little more work.


The last printer that failed on me under CUPS was actually a Brother one. I still haven't figured out why it worked for two days and then stopped; I guess networked printers are tricky.


Under OS X I have good success with brother printers... They usually work for months without fail and then stop working. The only thing that fixes it is to cycle power on the printer. I'm guessing buggy firmware on the printer itself.


Yes, absolutely! For getting distraction free focus and seeing the big picture of code, printing and highlighting pens are the best.

In the sense of good use of resources, printing 30 pages of mostly irrelevant slides for a meeting is not.

I do think, that not printing the page you really don't need makes some difference. Even if it is recycled, it is better not to cut down those trees in the first place.


> Printing 30 mostly irrelevant slides for a handful of people for a meeting is not good use of paper.

If they are irrelevant, yes.

But then again, printouts have their place in presentations as well. I recall Edward Tufte advocating for using handouts - i.e. if your talk features things like graphs, diagrams and data, it's a good idea to collect them into a document that you would then print and distribute throughout the audience. People can then look at the graphs in high resolution and also take notes.

Inspired by this, the last time I did a lecture on fixed timestamps in videogames I took all graphs, diagrams and source code fragments, put them on a single A4 page and printed a copy for each of 60+ people present. From what I could tell afterwards, this was really useful - people could consult examples in their own pace and there were zero complaints about source being unreadable for people in back seats. So the next time I'll be doing a lecture, I'll likely design and print handouts too.

Of course all of this applies to educational presentations, not to marketing ones. There's no point to waste paper on the latter, they're all about invoking emotions anyway.


> the last time I did a lecture on fixed timestamps in videogames I took all graphs, diagrams and source code fragments, put them on a single A4 page and printed a copy for each of 60+ people present

Not that you were looking for my approval, but this is 100% good use of paper in my books. Extra points for putting them nicely together in compact form, that all the relevant information can be seen with a single glimpse :)


> Extra points for putting them nicely together in compact form, that all the relevant information can be seen with a single glimpse :)

Yup. Usability + the fact that I really do hate waste.


I still keep a dot matrix printer and some fanfold paper around for printing code I'm working on - saves taping pages together.


We use printer accounting to get around this. Everything is billed to the relevant departments at an extortionate rate per sheet. The department has to pay the print bill out of their budget.

Printing has nearly stopped to the point that some departments have resigned their printers entirely.


It's interesting that this is a problem. I would guess that it's mostly to do with the problem of interaction and eye strain associated with on-screen information.

I can interact with most things on a monitor, but math is one of the things that I still need paper or a whiteboard for.

I have friends that can't stare at a screen for 16 hours a day like I can, and I assume that's the other problem.

Either way, I hate it when somebody comes by and has my email to them printed out and ready to "discuss" and the discussion is just yes or no.


To that end, I started insisting people only print things one-sided if they are going to give me something printed. It seems counterintuitive, everybody says "print double-sided to save paper". But the way I see it, that's wasting the WHOLE page, whereas single-sided wastes only half of the page, and I can write notes on the other side.

But yeah, it used to be so bad that I had to write a LOT of notes to keep up with it. Now, I just don't bother with offices at all. Nobody can print anything at me.


I agree with this sentiment. I do print code single-side (easier to work with, especially when you want to see all pages at once), and then use the other side for notes and design later. I keep the old printouts as scratch paper.


I know exactly what you're talking about. I have to deal with a lot of the "I have to print email to read it!" mentality at work. We started to track the printing and discovered that a building of approximately 1300 users was printing 100000 pages a month. The worst part is that every user has at least two devices (computer and phone, sometimes tablet) and access to the full Google Apps.


Technically you could take a photo of it, process it then chuck it into a recycling bin :).


Doesn't solve the waste though.


unmatched </semisarcasm> at (3,123)


Voted down? It's just a quick, weak joke, I don't see the harm.


"Resist complaining about being downmodded. It never does any good, and it makes boring reading."[1]

I imagine that a lot of people also find "quick, weak jokes" to make for boring reading.

[1]: http://news.ycombinator.com/newsguidelines.html


* TL;DR from http://www.phoronix.com/scan.php?page=news_item&px=MTgwMjE *

>> CUPS 2.0 has greater security around its scheduler, various OS X specific improvements, systemd support, support for TLS certificate validation and policy enforcement, updates to all Linux man pages, dropped OpenSSL in favor of GNU TLS, dropped support for AIX / HP-UX / OSF/1 architectures, and various other changes.


CUPS work very well for the company I work for.

In just one linux print server, we've got exactly 1110 printers installed (HP, Brother, Ricoh, Argox, Epson, Samsung and LexMark), all working without hassles.

We're a warehouse/retailing company.


"Dropped "dark wake" support on OS X, which was preventing portables from going to sleep when there was a stuck job. We now use a variation of the CUPS 1.4 sleep support to do a cleaner sleep (<rdar://problem/14323704>)"

At last.


I'm guessing this isn't a coincidence, with osx yosemite out as a release candidate?


It does seem to be somewhat OSX focussed - although dropping OpenSSL in favour of GNUTLS is going to be interesting, as currently I don't believe there's any sanctioned OSX build of GNUTLS.


I wonder if the OS X version also has SecureTransport support?


Don't forget CUPS is an Apple product. I'm sure they have a plan :)


Notably, the maintainer who is by far the primary author, is an Apple employee.


Not only that. Insofar as that is possible, Apple owns CUPS.

Look at the footer on cups.org:

CUPS, the CUPS logo, and OS X are trademarks of Apple Inc.


I'm on Beta 4, which I think is equal to RC, and I have 2.0.0 when I run:

cupsctl WebInterface=yes

and go to

http://localhost:631


I didn't thought about it, but you're probably spot on.


I find it interesting they dropped support for HP-UX, and AIX, and that there was support to drop for OSF/1


To be fair the number of HP-UX and AIX machines that need to be connected to printers can probably be counted on one hand.


I had more than a handful of AIX machines serving literally thousands of printers to deal with at a previous job. AIX was hosting a POS/ERP system which was heavily dependent on printing to trade (b2b/'trade' retail business) - it was an entirely business-critical function.

I never want to see another OKI ML520 again.


Yup.. http://localhost:631/

CUPS 2.0.0 CUPS is the standards-based, open source printing system developed by Apple Inc. for OS X® and other UNIX®-like operating systems.


Bought by Apple Inc. for OS X. It predates OS X by a few years.


CUPS 2.0.0 doesn't though.


Nice link! Works on my debian box (of course).


I wonder why there doesn't seem to be any implementation for AirPrint/ePrint in Linux. Seems like a good lightweight alternative to CUPS for wireless systems/devices.

Also, I am curious: is CUPS still used in OSX?


Certainly, go to http://127.0.0.1:631/ in your browser. CUPS is still actively developed by Apple.


I wish that the CUPS filter language would be made out of something like lua or javascript rather than binary (C) right now.

But I have less than zero hope of that ever happening - both OSX and Linux use CUPS as their printing framework. There is no way, Apple (which basically owns CUPS) will allow Linux to get easy leverage on the superb printer driver ecosystem of the Mac.

Before anybody else comments - printers do work on Linux. But not all features work.


It's been a while, but I thought that CUPS filters were just executables with an input stream and an output stream, so you could write them in whatever you wanted?


correct - but they are non portable. Which means that all the nice features that Mac drivers have are not available on Linux (or the printing is screwed up, etc.).

I wish there was some way to make portable/obfuscated filter drivers, so that vendors could ship them and be usable on all CUPS platforms.


printer are bloatware. Old men recipe: new corporate printer tends to have a lpd daemon running accepting postscript (60%). Use rlpr and send your file directly bypassing the spooler. nmap on ldp port my help.


We're going to miss you when you're gone.


I have no comment about CUPS, but I will say that I find this site to be pleasant on the eyes, even if it is just bootstrap. I feel like half the reason people don't want to use Debian or GNU stuff or other free software is because the sites themselves are fugly remnants of the mid-nineties and leave much to be desired from a usability standpoint.

How hard would it be to simply wrap their boring html and drop it into a bog standard bootstrap? Like, a day?


Why don't you volunteer to improve the design of the webpages of the OSS projects that you think would need it?


"Submit a patch."


By far not every project does it like php and host their website sourcecode on a github repo.


How do you know I haven't?


Funnily enough that page design is actually bugged on my WinXP/FF31.0. - the floating menubar part covers the top line of content. Looks like a screen-size problem (I'm on a 1024px laptop) that fixes with either smaller font-size or added padding to .breadcrumb.

My point I guess is that web design is often harder than it looks. OSS projects with limited resources are better using Github or Sourceforge or similar, IMO, rather than trying to keep on top of web page design decisions to cater for ever changing UAs.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: