Hacker News new | comments | show | ask | jobs | submit login

Warning: according to the known issues in the release log [0]:

* Skype won't start (without a fix [1])

* Google Chrome won't install (without a fix [2])

[0] https://wiki.ubuntu.com/RaringRingtail/ReleaseNotes

[1] https://bugs.launchpad.net/ubuntu/+source/skype/+bug/1155327

[2] https://code.google.com/p/chromium/issues/detail?id=226002#c...

"Actually, having a workaround by definition lessens the severity, as if someone wants to still run Skype, they can."

I don't think everyone is on the same page about Ubuntu being a consumer distro and not a hacker's only distro... I have definitely noticed a decrease in the stability of Ubuntu releases over the years and it's attitudes like this that seem to be in part responsible for it.


Oh I kept reading, this is better:

"Rainer, a better solution for international calls is to stop using Skype (which is closed source, and frequently has problems with its Linux client) and change to a service that supports standard protocols (SIP etc.), e.g. using Ekiga"

Yes, the client should change their software and infrastructure to better suit your product...

This makes it pretty much impossible for me to "sell" Ubuntu to any of my friends who aren't programmers AND tinkerers (many of my programers are too busy being productive to be interested in dicking around with making things work anymore, and increasingly I hold this view).

> many of my programers are too busy being productive to be interested in dicking around with making things work anymore, and increasingly I hold this view

This. I switched to OS X from Ubuntu about a few months ago simply because of this reason. I was too tired of tinkering and configuring endless .conf files, adjusting my GNOME/Xfce panel layouts for optimum productivity, and changing my display manager because all of them sucked.

In my final days with Linux I was running Ubuntu Minimal with twm -- I went completely old-school. Fewer things at the core, fewer things to go wrong.

I caved and bought a MacBook Air. While I miss the endless configurability of Linux, I find that most of the time I spend trying to configure OS X is only spent to make things more suitable for my workflow -- changing the number of desktops I have, changing the applications in my dock, or adjusting Exposé to show all app windows.

I never have to configure something because it breaks or isn't compatible with something else. I could have gone the FreeBSD way, but then, well, you know. OS X just works. As a Ruby and Java developer I find it perfect. I probably will try the 13.04 release on the Mac (I tried the beta and liked it), and I'll see how it goes.

It looks like a solid release, so I'd really like to keep it (but I probably won't because I'm cheap and only have the 120gig SSD).

Except for a few driver issues, (Nvidia and Optimus, doh!), I haven't had a issue with Ubuntu or Mint. Any other issues with configuration, I would have also had with Windows or OS X since they were application or server specific.

But I don't spend days tweaking the UI, I have a baseline of UI expectations that's not too difficult to meet. (But Windows 8 doesn't meet it... 'nother story for 'nother day...)

I've had tons of issues. Especially around power management, multi-monitor support etc. When I "Hibernate" I also need to roll a d10 to know if it'll actually come back up. This is anecdotal but on same hardware there are no issues under Windows and everything works as expected. Eclipse seems less stable and slower under Ubuntu which is probably related to the Java run time. Firefox also less stable. Webcam sometimes doesn't work. Audio suddenly stops playing until you touch the volume. Couldn't get WebEx working under Ubuntu (tried lots of suggested workarounds). The kind of stuff that just works in other OSes.

On what hardware?

I have the same issues on an Asus laptop. It was well supported on 12.04 LTS but the 12.10 is obviously a big failure. Not to speak about the system that slow down by itself, some swap mess I don't have time to fix, and the overall growing instability of the system tools. Memcheck segfault, then the report tool activate itself, inspect, segfault, launch again... I also upgraded a pc of a friend from an good old athlon xp to a new shiny i3, and guess what the official, libre, graphic driver is just unable to run correctly, apparently because of a bad switch in an internal opcode.

Something is wrong here. The massive shift of devs from linux to OSX should warn some people somewhere about the overall quality of their stuff. A quality that we have been proud for decades.

Dell laptop here (Core i7 + the dreaded Optimus)...

Have you checked out this? http://bumblebee-project.org/

my Canon Printer wont work properly in UBUNTU.

My printer, scanner, graphics tablet, and sound card won't work in windows (the printer manufacturer explicitly refuses to support 64-bit, the graphics tablet bluescreens win7 when plugged in, the others just do nothing). All work flawlessly out of the box on Ubuntu. Hooray for anecdotal evidence \o/

You don't like Ubuntu dictating a desktop environment to you. So you fiddle with other things for a long time, at your own option. Then you pay Apple to dictate a desktop environment to you. Hmm...

He explicitly said he misses the configurability, so I wouldn't blame him for that.

As a Mac user that runs triple boot, the "just works" mentality in OS X is present only as long as you don't step out of the typical use case scenarios. It's great to say be a writer on OS X; there are very good tools and apps to write in a minimal, esthetic environment (WriteRoom and others). On the other hand, as a "hacker", I've always had trouble in OS X -- from Page Up/Page Down not registering properly in Terminal, to full NTFS and EXT4 support (FUSE used to work, but I think it was not maintained) to getting vim, gcc and other tools of the trade working on OS X (yes, there are ports-like systems on OS X, but hardly ones that I'd say "just work").

On the other hand, in pure Linux world, brightness controls, backlighting controls, volume controls and multiple sound card support were all rather tricky when I tried using Gentoo on the same Mac for a time. Some of those I solved, but some I just gave up and was content that it kinda-but-not-completely works.

Ubuntu seems to be tested enough on Macs, so those things work on my laptop now without having to configure too much. It may not be the perfect "just works" experience, and I have seen a lot of things going the wrong way on Ubuntu, but currently it's a good mix of being easy to maintain and being easy to start doing actual work with.

I didn't leave the Linux world because I didn't like Ubuntu dictating a desktop environment to me. I left because OS X suits my needs better than any Linux distribution (except perhaps OpenSUSE) could.

Logical fallacy. You've built a strawman, based your argument off it, and then contradicted yourself.

No they didn't. Where?

It's strange because the overwhelming consensus so far has been that the latter releases have been much more tested and stable than any before, with the sole exception of LTS releases. Still, it's always been good advice to hold off on upgrading until a release has been out for a month or so, and for most users staying with the LTS release is probably good enough anyway.

There's a huge QA process for Ubuntu now -- packages have to go through a large swath of automated tests, and the alpha/beta versions are kept working on a daily basis. It was only 2 years ago or so that most Ubuntu developers didn't even expect the unstable release to be usable until the feature freeze milestone had well passed.

>many of my programers are too busy being productive to be interested in dicking around with making things work anymore, and increasingly I hold this view

This is a sign of maturity. Time is our one true finite resource, and we need to prioritize what we do with it.

I once bought a laptop which ran Windows 7. The Wifi did not work due to their drivers being broken.

The correct move then is to return it.

No. The correct move was to install Ubuntu, which had the correct drivers. Point being that all software has issues. More so if the software is as complicated as an OS with two releases per year.

You bought a product and it did not work as advertised. The recommended path would have been to have it serviced or return it to the distributor. You could also attempt to install the correct working drivers, as this most likely would not void your warranty. However, changing the entire OS is a very big not-recommended move on a recently purchased laptop that's having such a simple problem.

Why post something like this? I don't get it. He stated what he did to fix the problem and it apparently worked.

So his recommendation is proven to work and yours is....?

It's smothering the candle with a firehose. It's a very destructive solution for which avenues have been paved before-hand to solve such problems. Did it work? Yes. To fix one small problem (malfunctioning WiFi) he rendered himself incapable of doing other things (use Microsoft compatible software, and, no, I'm not including WINE).

In the end, if his purpose was to install Ubuntu, then good for him. However, I wouldn't give it as general advice out of voiding their warranty alone.

You missed the point by two blocks. Which is that OS have bugs. In my case, i was going to instalk ubuntu either way. I was making a counter argument against a comment above. Still, installing another OS does not void the warranty where i reside.

Guys, Skype is owned by Microsoft, and the Linux version hasn't been touched in like 2 years. I'm frankly not surprised it's b0rked.

So... yes, the user should use software that isn't deprecating support for their platform.

They are releasing steady updates for it. 4.1 came out Nov 2012. That's hardly 2 years. That's after the last Ubuntu release. And it's fully supported for the platform. What are you talking about?

Well that's news to me. The last I looked was before November, I guess, and before that it still had the same old crufty UI from like 5 years ago. The userbase had been pretty up-in-arms about it for a while too: http://community.skype.com/t5/Linux/Vote-here-for-linux-skyp...

Edit: also, they're still sitting on version 5, which introduced group video calling. It's a pretty sad situation overall.

Much as I can't stand to say something nice about Microsoft, the Skype updates at least on Ubuntu 12.04 have worked fine.

Ubuntu has no control over the Skype project. They aren't obligated to support every third-party product when it is impossible to do this.

I don't think this is the problem with recent Ubuntu releases.

Ubuntu has no control over the Skype project. They aren't obligated to support every third-party product when it is impossible to do this.

That's an interesting attitude.

Backwards compatibility is a generally regarded as a feature. If you read Raymond Chen's blog about the length's Microsoft went to in order to support old software on new versions of Windows you'll see how important some operating system vendors regard it.

Linux generally seems to have an interesting attitude to backwards compatibility. "It works on old hardware" is something that used to be regarded as a huge benefit of Linux.

These days - especially for Ubuntu - it doesn't seem to be regarded as a feature at all. Additionally Ubuntu seems to introduce breaking API changes in every release. I find that to be quite a surprising route to take for an OS that is trying to build desktop share.

I'd be quite interested to hear why people think they do this. Is backwards compatibility just not seen as important at all or is there a bigger strategy here?

Actually, they should try.

They are not obligated, but it is in their best interests to build a reputation of not breaking things.

I do think that Linux lacks a reputation of not breaking existing programs as much as it lacked a reputation of ease of use before Ubuntu.

Now, ease of use, and not breaking things? I bet that kind of reputation is worth millions of dollars.

Even Linus agrees about it, when it comes to the kernel breaking userland programs. The kernel developers have no control over userland programs. Nonetheless, the kernel developers try very hard not to break userland programs.

To be fair, I have definitely noted a decrease in stability of OSX releases of the last few years too.

At least they're consistent in not liking Microsoft.

Ubuntu bug #1, for reference https://bugs.launchpad.net/ubuntu/+bug/1

"Yes, the client should change their software and infrastructure to better suit your product..."

The person you quoted isn't even a developer as far as I can tell.

So, a "hacker's distro" should have bugs in it? No thanks, I'll take functional software.

I don't use Skype (imo.im works much better, and you don't have to install any programs), so I don't know anything about it, but Chrome seems to work.

Jitsi (jitsi.org) does video calling over Jabber and others.

Though it's not web based like imo.im, you also don't have to give your access credentials to a third party.

imo.im and similar services have one disadvantage: you give keys to all of your communications to additional 3rd party (as if microsoft wasn't enough)

3rd party? Microsoft own Skype.

You give your keys to imo.im.

They said additional 3rd-party, which implies MS is a third-party?

It is. Isn't it?

They aren't. They're the provider of the service.

...and a third party to your conversation with a friend.

Thanks for the plug, imo.im looks really cool! Can't believe I didn't know about it before.

It seems to be the spiritual successor to Meebo webchat. So far its the only web-chat site that I have found that supports Jabber. I would absolutely recommend it!

I use their Android app to stay signed into my professional and personal IM accounts at the same time. It's the best I've found.

imo.im is pretty great. I think they even had (web based) video calls with Skype accounts, at least until Skype restricted that type of access to 3rd parties.

Why can't I find a download link for linux anywhere?

Oh, there is no download for linux... I'm not much for browser chatting.

The Skype bug looks like a Multiarch problem and there appears to be a workaround in the bug report:

    LD_PRELOAD=/usr/lib/i386-linux-gnu/mesa/libGL.so.1 skype
which is trivial to implement either in a custom .desktop file dropped in ~/.local/share/applications or a custom launch script in /usr/local/bin (or whatever comes before /usr/bin in your path).

Trivial maybe if you're a power user that knows Linux and Gnome's desktop shortcuts by heart and isn't afraid of hacking, but definitely not for the regular user Ubuntu was intended for a long time ago.

Said user should then have a system administrator to set up the system. Really, you either read the manual to learn such things or you get someone to do them, expecting a system as complex as a computer to work out-of-the-box is nonsense.

What you seem to be saying is that in order to use a personal computer, you should be an expert.

I disagree, and I'm glad that most people disagree as well. User-friendliness is important regardless of what system it is, even Linux. Have you noticed that the most successful distributions of Linux are the ones that are most friendly to beginners?

This elitism is destructive and serves only to make the "experts" smug and keep everyone else in the dark. I've found that the real experts are the ones who are writing guides for solving issues so that beginners can solve them too. That way, everyone benefits... and the beginners get to learn too. Almost all of my experience with config files has come from having a problem, Googling it, copy-pasting some commands into the terminal, and then becoming interested and messing around with it.

I had a taste of those elitist attitudes a few years ago when I was looking up how to add a user on an Ubuntu server and the top result for a search on "ubuntu add user" was this page:


Ubuntu has two commands for this, useradd and adduser, where adduser is the easier one for newbies, while useradd is more flexible but has a lot of ways you can get it wrong. In fact, 'man useradd' recommends using adduser instead.

Unfortunately, that how-to page starts with useradd, and the first example on the page is:

useradd <username>

It isn't until later on the page that it mentions that there are some options you must add to the command if you want it to set up everything correctly for a typical user. And if you leave these options out, the page doesn't explain how to fix it after the fact.

Finally, near the bottom of the page it mentions the adduser command that would have worked with no special options and no hassles.

So, being at the time a Linux newbie, I worked through the how-to and tried the useradd command it suggested, got myself into a situation where I wasn't sure exactly what I'd really done and hadn't done, then finally discovered the adduser command I should have used in the first place. I posted a reasonably polite complaint that the how-to had led me down the wrong path and it ought to start with the recommended adduser command instead of just mentioning it in passing at the end.

The responses I got were fairly eye-opening. (Click the "show archived reader comments (36)" link to see them.)

This one was my favorite:

> Michael Greary [sic] you are a D&%K H#$&D. You’re the 1 that messed it up no1 else, and anyone else for that matter.

> Hopefully you’ve learnt by now that when blindly running linux commands you should read the whole article to make sure it’s what you want first.

> Blaming it on other ppl isn’t gonna help either. You should’ve said it by admitting your mistake, asking for a way to rectify it.

> (sigh) I bet you have no friends

I have to admit he was right. Clearly, I am not worthy of using Ubuntu!

That article is infuriating.

I searched for an alternative. The second Google result for 'add user ubuntu' is: https://help.ubuntu.com/community/AddUsersHowto

There are problems:

* The article has a pants title - no novice in their right mind is going to click the link over the first result with the better title

* It begins with a tag saying it's out of date

* It's marked as immutable so those who care can't make it better.

What can we do to make these terrible articles go away?

Of course, you tacitly skipped over the other helpful comments and instead chose to concentrate on the worst comment in that thread. However, that doesn’t render the message delivered there wrong in any way – read the whole article/how-to before starting, just as you read the whole recipe before starting to cook.

So, yeah, you

a) didn’t read the manuals supplied with your distribution

b) instead went directly to a random third party site via Google

c) blindly trusted said site without fully reading it

d) messed up while doing so.

You should have gotten a computer-literate friend to help you, just as you don’t fix your car by looking for an howto online and dismissing the manual you can usually find in the glove compartment.

Thanks (belatedly) for your comments.

The thing is, of course I know I screwed up. I don't need you or anyone else to tell me that. I already know it.

But that has nothing to do with my complaint: the top match for a Google search for "ubuntu add user" is a site that gives out poor quality information, doesn't improve that information as a result of user feedback, and has a community that is positively hostile to feedback. Yes, as you point out, there were also helpful comments in that thread. There were also several other people who had the same problem I did, and the site never did anything to improve its content.

Isn't it obvious that this was a lousy tutorial? Doesn't the author of a site like this have any responsibility to put out useful information? Does the fact that I screwed up make my feedback any less valuable? If you're writing tutorials, that's exactly the kind of feedback you ought to be looking for, so you can improve your tutorials.

Contrast it with Arch Linux's page on user management:


> To add a new user, use the useradd command:

> # useradd -m -g [initial_group] -G [additional_groups] -s [login_shell] [username]

Here they do use the same useradd command that I was complaining about (probably because Arch doesn't include adduser by default?), but they give a complete example with all the options required, followed by an explanation of those options.

Now that's how you do it.

BTW, I was almost tempted to take offense when you said that I "should have gotten a computer-literate friend to help". Yeah, I guess I'm not very computer-literate. After all, I've only programmed in Algol-60, APL, assembler (Sigma 7, PDP-10, 6502, TI 9900, 8080, x86), AWK, Basic, Bliss, C, C++, COBOL, CoffeeScript, Forth, Fortran, Go, Java, JavaScript, Jovial, Lisp, Lua, Pascal, Perl, PL/I, PL/M, Python, Ruby, SIMPL, SQL, Trac, and a few other languages you wouldn't have heard of.

But we're all newbies at something, some of the time. In my case it was basic Linux user setup.

Maybe you even screwed something up once, as a result of somebody giving you bad information that you didn't thoroughly check out. (Not saying you did, just possible.) If that ever happened, which would be a better response from the source of that information: chewing you out thoroughly, or correcting the bad information?

And this is what's so frustrating. I love Linux. I truly love the idea of making everything open from the kernel to the desktop environment to all of the applications you can run on it. I truly love the idea of customizing everything... and if you don't like what's there, you can modify it to your heart's content if you learn how. That's just freaking cool to me.

Except... we have douchebags in the community who talk shit to newbies and then cry, "Why does everyone use Windows? How come Linux isn't more widely used?"

Just check your post history and read the number of times you wrote "RTFM." That's why.

Well, there's shit-talking the newbies, and then there's refusing to lower your system to the lowest common denominator by dumbing it down.

Not being careful of that second one gets us things like the GNOME 3/Unity/Windows 8 design-by-committee that seems to be loathed so much by power users.

Is it so wrong to expect some basic minimum level of competence in a system before someone uses it? And then suggest that someone take the most basic of steps to acquire that competence before asking questions? (By R'ing TFM?)

Systems should as a rule be somewhat intuitive (IMNSHO), but you can only take that so far before you start hamstringing yourself. Devs would never get anything done if they spent all their time training people how to use systems.

You're definitely right that there is a fine line between user-friendliness and "The user is an idiot" in UI choices. Right now I'm using Lubuntu; I dislike Unity immensely.

Remember that everyone starts out incompetent. They ask dumb questions, and in some cases entitled newbies expect the experts to bend over backwards for them. But one thing that's nice is that a lot of questions have been answered before by patient experts.

Say there's a guy on the Ubuntu forums who asks the following question:

"Hey, how do I change my desktop resolution?"

An asshole answer is "Fucking Google it, idiot. It's not hard."

A decent answer is, "Here's a link to the answer. In the future, I suggest Googling your issue. You don't have to wait for us to answer it! Often, questions have been asked and answered before. We're here if you get stuck, though."

This has a few benefits. The first (obvious) benefit is that the newbie doesn't immediately think "Wow. What a dick, I was just asking a simple question."

The second benefit is that future newbies who Google the same question get a link to the answer.

Of course, there are lost causes (Where's the Start button? This isn't like Windows at all. I don't like it) and asshole newbies who simply refuse to try to learn anything by themselves, (Ok, you answered my question on desktop resolution. How do I change my desktop background?) but I think these people are the exception rather than the norm. And they can be dealt with in a way that is constructive rather than crass and uncouth.

Try giving the ‘decent’ answer to the thousandth person asking that very question without any sign whatsoever of the work they put in. Or to the person who obviously did not read the rules for the mailing list they signed up to and instead ‘just asks a simple question’. There are, unfortunately, many more entitled people than you think.

Fortunately, you can easily killfile people in mailing lists, and doing so quickly (i.e. after one wrong post) helped me greatly.

> What you seem to be saying is that in order to use a personal computer, you should be an expert.

No. What I am saying is that a) using a device is different from administrating it and b) in order to administrate a device you have to be an expert. You don’t expect to buy a car and never ever go to a workshop with it, because, you know, you shouldn’t have to be an expert to service your car. Similarly, expecting to be able to service a PC without being an expert is as futile.

> Have you noticed that the most successful distributions of Linux are the ones that are most friendly to beginners?

What metric/definition of successful are you using? [0]

> (third paragraph)

Exactly. Your point being?

[0] Hint: Number of users is not a measure of quality, and hence not necessarily a measure of success.

I would argue that popularity is actually a great indicator of the quality of an OS and your experience with it.

I am what you would consider an "expert" (or at least, I can definitely competently administer a linux computer) but I don't use linux distributions for day-to-day computing. The reason is that linux computers, even ones running "popular distros" such as Ubuntu, are considerably more prone to interoperability foibles and installation/updating problems.

The reason popularity matters, even though I have the technical skill to administer any OS competently, is that popularity is the single best incentive to encourage third parties to iron out the kinks on a particular platform. The reason AMD drivers work better on Windows than they do on Linux is not because Windows is "better" by any of the metrics of quality you espouse, it's because Windows is more popular. The reason Steam for Linux is only supported on Ubuntu is not because Ubuntu is "better" it is because it's the most popular distro. The reason to use popular platforms is because all of the little things that require a half-hour of expert time to work around would have already been solved by the vendor who recognized that such issues are worth testing for and fixing for a platform with 100 million users, but not for one with 20,000. That's distasteful to some because it doesn't seem meritocratic and it's out of the control of the creators of the OS, but it's a fact of business, and it does affect the quality of an OS in tangible ways.

As for the "expert" thing, the costs really aren't that different for various kinds of users. The reason I care about this stuff is because the opportunity cost of my time spent solving menial issues that would have been solved by the vendor if I were on a more popular platform is too high. The reason my mother cares about this stuff is because the transaction costs of finding someone who will spend the time to figure out what's going on and fix her computer is too high.

> The reason popularity matters, even though I have the technical skill to administer any OS competently, is that popularity is the single best incentive to encourage third parties to iron out the kinks on a particular platform.

I prefer a working platform that forces me to read the manual beforehand over a platform where I don’t have to read the manual but random third parties iron out kinks.

Popularity is a measure of quality, but only of quality wrt to the usecases of the people among which the product is popular. My computer usage is very different from my mum’s, and likely from 95% of the population, hence, the popularity of a product among 95% of the population is rather irrelevant to me.

If you’re a single person with a tight budget looking for a car, you don’t look at car popularity among large families owning oil fields, but among people similar (in the relevant ways) to yourself.

Hence, popularity among a large user base is an indicator that the OS is optimised for use by a large part of the population. If you find that this is indicative of the quality you’re looking for, fine.

> I would argue that popularity is actually a great indicator of the quality of an OS

Windows is tons better than OS X then?

Windows, is widely distributed...not popular.

Couldn't you say the same about Ubuntu? It's far more widely marketed to end users than any other distro.

No, Ubuntu is chosen by its wise users, not "marketed".

So wise.

I was specifically using "successful" to mean "number of users." People use what they like, and Ubuntu has gotten a lot more users than, say, Gentoo. Is Ubuntu superior to Gentoo? That's a matter of opinion; I'm sure that there are plenty of applications where people have decided that Gentoo is the perfect fit for what they want to accomplish. But for personal computing needs, people tend to use what is most user-friendly.

I don't mean that it's okay to be completely illiterate in how a computer works, especially if you're using Linux. To take your car example, I don't think that a driver should know how to rebuild his engine. I do expect the driver to know how to change his oil and brake pads and know about scheduled maintenance.

The same is true for computers. A Linux user should know basic bash commands, how to install software, how to research problems, and so on. But he shouldn't be expected to know how to troubleshoot driver problems or know the ins and outs of xorg.conf. That's what Google and the aforementioned benevolent experts writing guides are for.

While I agree that we should do everything we can to be user-friendly and not allow what I'm about to suggest to take the form of "elitism", I think we underplay the importance of helping the user understand the system. I think your typical user should understand SSL before they do online banking and shared libraries before they let their work rely entirely on their computer, in the same sense that I think you should know how to change a tire and change your oil before you buy a car or go on a long road trip.

Yes, they all should just "work out of the box", but in a more and more literal sense we trust a lot of machine with our lives. We ought to understand all of these machines better before we trust them with our lives or vote on the leaders who will regulate their use.

We have abstraction layers precisely so that you can use things without understanding them. It's not just computers, every day we use all kinds of technology without properly knowing how it works.

Of course, sometimes those abstractions are leaky, and we have to change a tire or apply an update. But suggesting that typical users should read up on SSL (and therefore asymmetric cryptography) before they use online banking is as ridiculous as saying that every new driver should study the detailed workings of the internal combustion engine. Someone needs to know how it works, but my mother really doesn't.

I've personally seen people click through the most alarming-looking certificate errors and start entering their bank login credentials. I've driven in cars with people who had put up with that knocking sound in the engine for months because it wasn't getting any worse. If your mother did any of these things, I would be very worried about her financial and physical safety.

You don't have to know how either works to recognise something is wrong with them. My mother, happily, is quite cautious and would go and find someone who understands the machine better. In the computer case, I'd likely be getting a phone call.

The certificate errors are kind of a design failure. Users are so used to clicking through messages they don't really understand, and this is just one more. Even the most alarming certificate errors are usually innocuous, in my experience - someone forgets to renew a certificate, or fails to get a new one for a domain change. That doesn't mean it's OK to ignore them, but we really need systems that don't cry wolf unless there's a really good chance that there's a wolf.

> Even the most alarming certificate errors are usually innocuous, in my experience - someone forgets to renew a certificate, or fails to get a new one for a domain change. That doesn't mean it's OK to ignore them, but we really need systems that don't cry wolf unless there's a really good chance that there's a wolf.

How do you propose to algorithmically determine if a certificate not matching a domain change is innocuous? This is a bit like suggesting having a "check engine" light that only comes on when there's an immediate risk of an engine fire.

For starters, there are a whole load of conditions that are no less secure than a standard http connection, such as self-signed certificates. But in those cases, the browser throws up a huge and misleading warning about it being insecure. Insecure is the default on the web: we should focus on positive signs that a site is secure.

Secondly, the system should make it harder for the administrator to make mistakes. For instance, the web server could refuse to serve https on a domain that it didn't have a certificate for. Or when a certificate nears its expiry date, the administrator should be getting plenty of reminders about it.

Thinking bigger, what if we tied it closer to the sensitive UI? What if only pages loaded securely could show a password field? What if rather than typing credit card numbers into boxes on a webpage, we were used to using a special browser interface that would only light up on secure pages?

Malicious intent can be algorithmically determined by checking for packets with the "evil" bit set to 1, as specified in RFC 3514 [1].

[1] http://www.ietf.org/rfc/rfc3514.txt

That is a horrible specification.

"Multi-level insecure operating systems may have special levels for attack programs; the evil bit MUST be set by default on packets emanating from programs running at such levels. However, the system MAY provide an API to allow it to be cleared for non-malicious activity by users who normally engage in attack behavior."

This should say that the system MUST provide such an api. Otherwise, a you can find a situation where the reciever may have to allow evil packets to pass, because they are sent by a user who has no option of setting the evil bit to 0 for non-malicious activity. Kind of defeats the point.

"Fragments that by themselves are dangerous MUST have the evil bit set. If a packet with the evil bit set is fragmented by an intermediate router and the fragments themselves are not dangerous, the evil bit MUST be cleared in the fragments, and MUST be turned back on in the reassembled packet."

This places an undue burden on intermediate routers, as they now have to parse the packets and determine if a fragment is evil.

"Intermediate systems are sometimes used to launder attack connections. Packets to such systems that are intended to be relayed to a target SHOULD have the evil bit set." These packets are not 'evil' in any technical sense. They are only a request for another computer to send evil packets.

"In networks protected by firewalls, it is axiomatic that all attackers are on the outside of the firewall. Therefore, hosts inside the firewall MUST NOT set the evil bit on any packets."

This is just stupid. If I am attempting to send malicious packets from within the firewall, I am already required to set the evil bit. Now I am also required not to. If I am sending a non malicious packet, I am already required to set the evil bit to 0.

Also, If I am on a computer behind a firewall, and attempt to send malicious packets to a computer outside of my firewall, I am still forbidden by this to set the evil bit. This will force attackers to either, break the spec, or give up their own firewall.

"Because NAT [RFC3022] boxes modify packets, they SHOULD set the evil bit on such packets. "Transparent" http and email proxies SHOULD set the evil bit on their reply packets to the innocent client host." I must be mis-understanding this one, becuase non of those uses seems evil.

"Devices such as firewalls MUST drop all inbound packets that have the evil bit set. Packets with the evil bit off MUST NOT be dropped. Dropped packets SHOULD be noted in the appropriate MIB variable." Dropping packets is a technichal inevitability, and a crucial part of TCP's anti-congestion stragety.

Just becuase the evil bit sounds like a good idea, does not mean we should go ahead and implement it without carfully looking at the technical implecations of the spec.

The problem with the car analogy, as TallGuyShort points out, is that cars are not capable of having as strong an influence over our lives as computers are. When I think about how this used to be done before, usually finances were done over phone where people got to talk to other people who explained to them things that they did not understand.

I agree that learning about SSL is probably overkill; but we do need more computer literacy among the general population. Think of it like this: finance is something that is hard, yet people learn about the basics so that they can keep their money safe. Basic computer literacy should have a similar status as a skill that is required to, say, keep your information safe.

I think the analogy is better than you are giving credit. I would say cars can in fact have much stronger influences in our lives than online banking can. For example, cars are much more likely to kill you. It's a much more important skill to recognize when your brake pedal is getting squishy so that it doesn't fail when you are careening down a mountain road, or recognize that an oncoming car is swerving erratically, than it is to realize you are logging in to an insecure phishing site. In the former case you end up dead, in the latter case your checking account gets wiped.

To clarify, when I say "learn about SSL", I mean people should know how to tell if the connection is "secure" or not, and that secure ensures that there is an extremely high probability that the domain they are talking to is who they say they are, as verified by some people that Google / Microsoft / Mozilla decided were trustworthy. Similarly, I think people should understand that emails can travel unencrypted between sites and are easily spoofed. I'm not saying they have a working knowledge of the RFCs for SSL and SMTP. I agree with the way you worded it - perhaps I made it sound too involved.

edit: another, perhaps more humorous example. How many times have you heard about people "hacking" their Facebook accounts and getting "viruses" on Facebook. I've been asked by multiple people to fix their computers, when the real problem was they had no idea what they were doing on the Internet.

So, why should I use Linux rather than Windows, which will run all my software? Expecting everything to work perfectly may be nonsense but I have to assume that your average user is going to follow the least effort path.

I never said that you should use Linux or Windows or Mac OS or iOS or whatever else you come up with.

Use the product you feel is best suited to your task. If that is Windows, use Windows.

Edit: That said, you will likely want someone knowledgeable around when setting up Windows or Mac OS X, too, so I don’t really get your point.

The last version I had any trouble with, by way of Windows, was 98 on a Tiny computer I had as a kid, mostly because of esoteric driver requirements. After that, it pretty much just worked. What do you think your average user is likely to do to it that would need them to go whistle for a sysadmin?

I don't know about anyone else, but I still go looking for an expert when I need any of my Windows machines fixed: whereas when any of my Ubuntu machines have issues, I'm comfortable and motivated to look up the answers myself. Partly that's an issue with documentation quality - you're a lot less likely to catch a drive-by download or follow maliciously bad advice on Linux forums - but I also find the workings of Linux much easier to understand.

"help with windows", 2.5 Billion, with a "B", hits. Do not pretend 100% or even a large percentage of Windows users never need power user help at a minimum.

The most popular desktop operating system on earth returns 2.5 billion hits. So what? Large numbers without basis for comparison are meaningless. I can't think how you'd get that basis here. The only thing you can really say on that basis is that less than 100% of users had no problems.


I didn't pretend that people never needed help with Windows. I said I hadn't had any real problems with it since 98 and asked what he thought the average user needed a sysadmin for in Windows.

I can see how you'd come to the conclusion you seem to have, but I really don't appreciate being made out to be a liar. If you act on a similar assumption in a similar manner in the future then I'm going to ignore anything else you have to say.

What does seem relevant to me is that there's a dramatic difference between an operating system that even programmers get tired of trying to make do what they want and jump ship from and one that you can reasonably expect not run into significant problems with for years. If Windows were equally messed up as Linux I'd expect to run into the problems at around the same rate.

How long does it take you to run into trouble with Linux? I tried it just now, downloaded Mint and threw it in the old DVD drive - 15 minutes. Most of which were spent trying to work out how to change the primary monitor so I could get the taskbar onto the one I wanted. When I did manage to do so the graphics corrupted and the system froze irrecoverably.

You clearly aren't the "computer whiz" in your family, who has to field all the "can you help me? Microsoft isn't working. I think I've been hacked, do you think it's a virus? Why is my computer so slow? Why is fixing it taking so long?" issues.

As for your ubuntu issues - you booted off a liveCD. Good for you. Now, how many liveCDs have you seen of Windows? How widespread are they?

Anyway, the same problems apply in setting up the system regardless of Windows or *nix - you need a power user to get you through the humps. Good luck installing XP as a normal user when you run into the ever-present problem of no appropriate network drivers. Basically you're comparing apples and oranges.

> You clearly aren't the "computer whiz" in your family, who has to field all the "can you help me? Microsoft isn't working. I think I've been hacked, do you think it's a virus? Why is my computer so slow? Why is fixing it taking so long?" issues.

There's a limit of naivety beyond which everyone will need help, yes. I simply believe that you can get away with being more naive with Windows most of the time. I don't have to fix my mother's Windows problems, or my sister's - from time to time I have to fix my father's, but even then not just for setting the thing up.

> As for your ubuntu issues - you booted off a liveCD. Good for you. Now, how many liveCDs have you seen of Windows? How widespread are they?

I think I've seen two or three versions over the years, mostly as repair tools environs. Why?

> Anyway, the same problems apply in setting up the system regardless of Windows or *nix - you need a power user to get you through the humps. Good luck installing XP as a normal user when you run into the ever-present problem of no appropriate network drivers. Basically you're comparing apples and oranges.

Windows Xp network drivers - that's three versions out of date. And really, where are your discs? If you got it ready made they should have sent you the drivers along with their bloatware and if you didn't then they should save sent you the mobo drivers with the mobo.

To be fair, Windows 7 typically still has that network driver problem with new ThinkPads. Fortunately, ThinkPad is a good company, and they make it relatively easy to locate and download the right network driver to put on a thumb drive, if you don't have an install disk handy (happens more commonly than you suggest). This is easy if you know what you're doing, but, of course, next to impossible without some power user knowledge (it also requires access to another computer with a working internet connection and a thumb drive).

And also to be fair, XP is in a valid sense only one version out of date. Vista was a fiasco, and that's not just a meme -- Vista was released way ahead of schedule and still had a ton of known unresolved bugs. It was by far MS's most buggy release, and the whole thing was a disaster. Compared to the relative stability of XP, Vista was a step backwards. And while Windows 7 can rightly be called a newer version of Windows XP (and is better in nearly every way), it would be a stretch to say that Windows 8 is a newer version of Windows 7. Windows 8 is a fork into a different paradigm, and is so different it's not even comparable to Windows 7 along a number of dimensions.

All that said, XP is still pretty damn out of date, even with the latest service pack installed, and if Windows 7 is an option, it's usually a better one. Of course, XP uses less memory, so it may be suitable for particularly old computers. And for people that don't like to pirate software, owning a copy XP is a great reason to stick with XP (if it works well enough, why fork out the money for Windows 7?).

"If Windows were equally messed up as Linux I'd expect to run into the problems at around the same rate."

My personal experience (as someone who has been dual-booting Windows and Ubuntu for nearly a decade and uses both operating systems regularly) is that both have a lot of annoying problems, but Windows has always been way more of a pain in the ass, and I've run into more problems using it. Especially when it comes to driver issues, Windows has problems (it took hours to get a standard WiFi card to work, and once my video driver got corrupted so badly the system wouldn't even let my try to fix it, and restore failed -- I had to reinstall the OS to fix it).

Will this be fixed in an upcoming ubuntu or skype update?

Both bugs appear to be things which a future ubuntu or chrome/skype update should fix. Please correct me if I am wrong.

Linux haters can't stand that Ubuntu can release new OS versions so often. OTOH, I'm tired of the chase and from now on will just stick with the LTS releases. If I were a some years younger though, I'd be there in the thick of every release. Kudos to the makers and the doers of Ubuntu. RedHat, where were/are you.

I have noticed that Ubuntu haters are not Linux haters at all, they are previous Ubuntu users, or simply fans of other distributions.

P.S.: RedHat is selling server versions and support for millions of dollars. Ubuntu still is not profitable despite all they have done for the community. It's unfair to me.

I dunno what happens if you install them after you upgreade, but google-chrome and chromium still work for me, as I installed them long before this upgrade. :)

With chrome it is just in packaging problem - thus chrome is not open source nobody can fix it on their own.

I'm using Chromium from the Ubuntu repository, which is good enough for me until they get it sorted out.

Sounds like you could also use the dev channel Chrome.

I've upgrade a couple of days ago. Skype works fine, I'm using version, not sure if it was upgraded during the process.

Chromium works fine.

Regarding google chrome it has been fixed on the unstable channel for a while now.

This is why I don't use Ubuntu anymore. Every single release they roll out has serious issues interrupting a normal user's workflow.

Ubuntu, if you're reading this: please consider testing your releases more thoroughly.

Maybe Canonical is releasing bugs on purpose to compete with Microsoft.

VMware tools will also not install.

any knows fixes for spotify ?

I don't recall needing any fixes for it, but it crashed randomly enough (and not just on Ubuntu) that I switched to their web client.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact