How To Create an LVM Drive

I posted some time ago about how to add a new physical volume to an existing logical volume, but I need a new logical volume now. So let's write down the steps for doing that.

1. Add the physical drive to the machine, and then format whatever space you want to use on it. Nowadays, I use gparted to do this, as fdisk doesn't seem to support really big drives. So we need to choose "lvm2 pv" (LVM version 2, physical volume) as the type.
   Let's suppose that partition is /dev/sdf1.
2. Create the physical volume.
   
   > pvcreate -t /dev/sdf1
     TEST MODE: Metadata will NOT be updated and volumes will not be (de)activated.
     Physical volume "/dev/sdf1" successfully created
   > pvcreate /dev/sdf1
     Physical volume "/dev/sdf1" successfully created
   
   I always use the "-t" option first, to make sure the command will do what I wish.
3. Check your work, if you wish:
   > pvdisplay /dev/sdf1For those new to LVM, the "physical volume" is the closest layer to the hardware. It represents actual physical diskspace.
4. You can repeat those commands to create more physical volumes if you wish. These can then be grouped together in a "volume group", which basically acts as if it is one huge disk, even though it may physically be spread across a number of devices
5. Create a volume group in which our new physical volume(s) can reside:
   > vgcreate newvg /dev/sdf1 [/dev/sdg1 ...]
     Volume group "newvg" successfully created
6. Check your work, if you wish:
   > vgdisplay newvg
7. Create the logical volume:
   > lvcreate -n newlv -l 100%FREE newvg
   The -n option gives the volume a name, and the -l option says how much space in the newvg volume group to allocate. There are a lot of ways to give this option. See the lvcreate manpage for info.
8. Check your work:
   > lvcreate newlv
   This will also show you the path to the device. Here on Fedora 20, it is: /dev/newvg/newlv. I.e.: /dev/VolumeGroup/LogicalVolume. I don't know if that can vary from distro to distro.
9. Finally, we need to create the actual filesystem:
   > mke2fs -L NewVolume /dev/newvg/newlv
   Of course, there are other ways to do this, depending upon exactly what you want to do.

The LVM HowTo has lots more information.

Great Series of Articles on Men and Sex Work

Just reading a fantastic series of articles by Catherine 'Chas' Scott on sex work. Find them here:

• http://allthatchas.blogspot.co.uk/2012/11/where-do-i-stand-feminist-pinball.html
• http://allthatchas.blogspot.co.uk/2014/02/melissa-gira-grants-playing-whore.html
• http://www.allthatchas.blogspot.co.uk/2014/04/katha-pollitt-melissa-gira-grant-and.html 

But perhaps the most interesting of these posts is this one:

• http://allthatchas.blogspot.com/2014/05/men-fantasies-and-nancy-friday-thoughts.html

which reflects on Nancy Friday's classic study of men's sexual fantasies, Men In Love. As Scott notes, what's striking about that book, published in 1980, is how little men's fantasies seem to have changed, even with all else that has.

Setting Up a Private Git Repo

For some time now, I've been running my own git repository on a server to which I have ssh access (since it's mine). Mostly, this hosts copies of various websites I run, such as my own personal site and various course-related sites. It makes it easy for me to be able to update the sites from different machines without things getting out of sync.

Details for how to do this are here:
  http://git-scm.com/book/en/Git-on-the-Server-Setting-Up-the-Server
But since I'm the only user, I didn't need to set up a special git user, as described there. The point of doing that is to allow multiple people to access the repo (though you could also manage that just with file permissions, if they all have shell access anyway). The only thing I actually had to do, then, was to set up the repositories on the server and commit stuff to them.

Since I occasionally need to create new repos, here are the steps.

On the server:

# cd /git
# mkdir /newrepo
# cd newrepo
# git --bare init 

It actually doesn't matter where you put the repos. But putting them in some central location makes a lot of sense, right?

Then, on the client where I've got the stuff I want to put in that repo:

# cd /path/to/local/
# git init
# git add <files>
# git commit -a -m "Initial commit."
# git remote add origin myserver.com:/git/newrepo
# git push --set-upstream origin master

On various other clients, I can then clone the new repo as usual:

# git clone ssh://myserver.com/git/newrepo

 Of course, this doesn't allow anyone else access, even read-only access. But since it's just for me....

Extract Pages From a PDF

Another note to self, this one concerning how to extract particular pages from a PDF:

pdftk infile.pdf cat pages output outfile.pdf

 Note that "pages" can be as complex as you like, e.g.: 1 2 5-10.

Add a Password to a PDF

Silly reminder for self. To add a password to a PDF:

pdftk infile.pdf output outfile.pdf userpw password

Obviously, you need to install pdftk to do this.


Many places you see online say it has to be "user_pw", which is also what the pdftk manpage says, but the underscore appears to be optional.

Recent Paper: Intuition and the Substitution Argument

I'm not sure why it never occurred to me to post announcements of new papers here, but, well, better late than never.

This paper, "Intuition and the Substitution Argument" (PDF here), was delivered at the Analytic Philosophy symposium at the University of Texas in early December, and before that at Duke University, in October. It will appear in a special issue of Analytic Philosophy also containing the other papers from the symposium, by Mike Martin, Tamar Shapiro, and Ralph Wedgwood.

Abstract:

The 'substitution argument' purports to demonstrate the falsity of Russellian accounts of belief-ascription by observing that, e.g., these two sentences:

(LC) Lois believes that Clark can fly.
(LS) Lois believes that Superman can fly.

could have different truth-values. But what is the basis for that claim? It seems widely to be supposed, especially by Russellians, that it is simply an 'intuition', one that could then be 'explained away'. And this supposition plays an especially important role in Jennifer Saul's defense of Russellianism, based upon the existence of an allegedly similar contrast between these two sentences:

(PC) Superman is more popular than Clark.
(PS) Superman is more popular than Superman.

The latter contrast looks pragmatic. But then, Saul asks, why shouldn't we then say the same about the former?

The answer to this question is that the two cases simply are not similar. In the case of (PC) and (PS), we have only the facts that these strike us differently, and that people will sometimes say things like (PC), whereas they will never say things like (PS). By contrast, there is an argument to be given that (LS) can be true even if (LC) is false, and this argument does not appeal to anyone's 'intuitions'.

The main goal of the paper is to present such a version of the substitution argument, building upon the treatment of the Fregean argument against Russellian accounts of belief itself in "Solving Frege's Puzzle". A subsidiary goal is to contribute to the growing literature arguing that 'intuitions' simply do not play the sort of role in philosophical inquiry that so-called 'experimental philosophers' have supposed they do.

Thanks a ton to David Sosa for inviting me to the symposium, and to everyone there for showing me way too good a time.

New Paper: Is Frege's Definition of the Ancestral Correct?

I've posted a new paper to my website, titled "Is Frege's Definition of the Ancestral Correct?" (PDF here) The paper is scheduled to appear in a special issue of Philosophia Mathematica edited by Roy Cook and Erick Rech.
Abstract:

Why should one think that Frege's definition of the ancestral is correct? It can be proven to be extensionally correct, but the argument uses arithmetical induction, and that fact might seem to undermine Frege's claim to have justified induction in purely logical terms—a worry that goes back to Bruno Kerry and Henri Poincaré. In this paper, I discuss such circularity objections and then offer a new definition of the ancestral, one that is intended to be intensionally correct; its extensional correctness then follows without proof. It can then be proven to be equivalent to Frege's definition, without any use of arithmetical induction. This constitutes a proof that Frege's definition is extensionally correct that does not make any use of arithmetical induction, thus answering the circularity objections.

In the general case, the new definition is fairly complicated. But in the special case of the concept of natural number, it reduces to:

n is a natural number iff there exists a Dedekind finite concept (or set) F such that F0, Fn, and ∀x∀y[Fx & Pxy & x≠n → Fy]

The last condition says that F is closed under successors, except that it need not be true of the successor of n. The point, which has also been noted (independently) by Aldo Antonelli and Albert Visser, is that Dedekind finitude can be used here, so the definition is non-circular. The intuitive idea is just that, if n is not finite, then the other three conditions force every natural number to be F, in which case of course F is Dedekind infinite. If n is finite, by contrast, then F is just [x: 0 ≤ x ≤ n].
One can then go onto prove induction from this definition, using many of the results Frege proves for his own definition of the ancestral.