* No transport security. As many people mention, at least adding HTTPS would help with this. However, most non-browser SSL clients (wget, curl) don't include any root certs by default so even switching to SSL would not help this method. Firesheep, sslstrip, etc. automatically generate a self-signed cert which would look no different to wget than a real cert.
* No persistence. If you download any installer package once and then reuse it on multiple machines, you get the benefit of knowing that the same code was installed on each machine (good or bad). With this method, users may catch the site in the middle of an update and get multiple versions of the package.
* No authentication. Even with SSL, you only get strong transport security. You would know strongly that ".pow.cx" sent you some code, but not how that code got put on the server. With package-signing, typically done on the developer's end system, you know that it was protected even before it was uploaded to some site.
* Easier to trojan than binaries. Inserting a few extra shell commands in a single HTTP(S) session (say, targeting a single client IP) is much easier than building a custom binary package. Consider how hard it is to even compile Firefox with all the dependencies. Now do that work and insert a trojan and upload a separate 10 MB binary that needs to be stored somewhere on the server while waiting for that one client to visit the site. Compare this to keeping a two-line patch to a shell script (easily done in RAM, maybe even by hotpatch).
* Trains users that all the above is ok since the popularity of this "| sh" install method is relatively new. (Yes, I know about shar scripts in the past but those ended by 1996 or so with the advent of real package managers). It is absolutely impossible to retrofit "| sh" to be secure, whereas it is definitely feasible to add package signature verification support to gem or yum or apt or whatever (in fact, all those already support it).
The fact that many installers aren't signed today is not an ok to drive this process back to the 80's. We should be moving toward the future when package signing is a required part of being a software developer. Too hard? Well build tools to make this easier!
EDIT: Okay, I see it's because of the use of sudo. But graphical installers often require the root/administrator password, and could be equally destructive.
curl https://get.pow.cx/ | sh
Fix your complaint? Like the grandparent said, I'm not sure why curl | sh is any less secure than gem install or whathaveyou, in the oh-god-this-script-just-ran-rm-rf-/ sense.How does this practice confound security more than a normal installer? sudo asks for my password just as a normal installer would.
Oddly, I'm more used to seeing arguments that distributing source code is better than distributing binaries because you can inspect source code.
We're arguing levels of badness here so it's a little hokey. But if you decide to open up your machine to run arbitrary code, a machine that can run shell will arguably get more infections than one that runs executables. To infect the ladder any script kiddies will need to know a 'harder' language and at least how to compile it. It's a couple more hoops to jump through. In the other case I could drive by and do scp ~/mailbox me@myserver:
We're already talking about running arbitrary code on a machine, compiled versus interpreted is irrelevant. And I think you have forgotten that a script with the appropriate hash-bang and file permissions is indistinguishable to most users from a compiled executable.
curl https://github.com/37signals/pow/raw/master/install.sh | sh
You can for example, at your network level point get.pow.cx to a malicious script and you're done. That's the security issue, it has nothing to do with the HTTP protocol.
With that being said, I don't care, the risk is the same as downloading any software via http, in fact I loved it, so easy :-).
Those that don't care won't look at the script any more than they'll check the md5 hash of a binary to see that it's a legit binary. For those that care, they can look at the bash source.
Since you are the security expert at Hn, we are trying to understand/learn from you.
This is not plain Nerdgasm making people understand about software security is making the world a little better.
The installation process is short and fully documented: http://get.pow.cx/
The web site and manual encourage you to read it.
I think it's far more transparent than, say, an OS X Installer package.
I think you would be doing the universe a small but meaningful favor not to advertise this installation mechanism.
But it is a very cool tool and a really well-done site. Congrats!
I'm not seeing how Pow's installation process is any less secure than, say, downloading a disk image from a random site.
With things like this that come from reputable sources, it's not unreasonable to put some trust in the source and some trust in the smaller percentage of developers who actually read the source code.
Even so, assuming you want to be extra careful (you're installing it on a development machine with highly sensitive information), it's not that difficult to clone the repo and edit the install script to install from your local copy of the repo.
Although I suppose it is possible that 37signals' GitHub account was hacked and someone maliciously designed a convenient Rack server and website with the intent of targeting the lucrative Rails developer demographic, or that the package the installer downloads is an insidious 37signals trojan not built from the code at the public repo.
Point taken.
It seems stupid, but on the other hand I never see people abusing the sticky bit, possibly as a result of the higher barrier to entry.
On the other hand, it causes me inconvenience every time I want to do something legitimate with it.
It is not really a script vs. binary issue so much as protecting against how scripts are loaded.
curl_review() {
result=`curl "$@"`
echo "$result" > `tty`
echo "Enter to proceed, Ctrl-C to abort:" > `tty`
read
echo "$result"
}
Then just change curl to curl_review: curl_review get.pow.cx | sh
Of course this kind of defeats the purpose of these easy installation tricks.cUrl is a shitty way to install software.
This is an idea that I think started with Ximian back in 2000 and I think we're ready for it to die. It'd be neat if the authors of Pow were cool enough to strike it from their (otherwise amazing) front page.
(I'd also be happier if the thread where the guy explains how Pow works and what it's components are were voted higher than this comment.)