It's always been hard to see distant planets because they don't emit light. Hubble can see distant galaxies because they contain lots of luminous objects.
Someone feel free to correct me if I'm wrong, but it doesn't seem that XGrid offers anywhere near the flexibility and power of Sun Grid Engine, which is open source and currently supports OS X among many other Unix platforms. SGE allows for tight integration of MPI processes, works for heterogeneous and distributed clusters, has lots of configurable options for remote monitoring, prioritization, and submission of jobs, and works well from the shell. The main selling point of XGrid appears to be the ease of configuration, which makes a difference particularly if your cluster of machines is "fluid" rather than fixed.
This class is in the Stanford CS193 series, which consists of lighter workload classes often taken by students outside the CS department, and geared for specific technologies. I don't believe CS majors get much credit for taking these classes. The core CS classes at Stanford are much less focused on details of specific languages (although Java is featured in the main intro OOP class; I guess you have to pick something).
Stanford has offered classes in.NET, Microsoft MFC, and Java, so a class in Cocoa is not too surprising.
As another poster mentioned, Ctrl-V in vim has allowed rectangular selection for some time. In emacs, cua-mode allows rectangular selection as well, and it's easy to use.
I've heard nice things about BBEdit, but if you use several platforms regularly (not just mac), I find a cross-platform editor is preferable. Besides, it's unlikely that BBEdit will ever be able to support such specialty needs as a mode for the IDL language.
We run a lab with 5 XServes running OS X Server (10.2). On the one hand, Netinfo has been a convenient way of sharing user accounts across the XServes. We store user accounts, NFS exports, and DHCP configuration information in the head node and it seems to work ok. DNS stuff is configured with traditional BIND text files.
On the other hand, twice during configuration we managed to corrupt the root Netinfo database on the head node. Once we were able to recover from a backup; the other time we corrupted it so badly that we could no longer log into the machine and had to reboot into single user mode and do some nontrivial recovery.
Granted, we ran into trouble mainly because we were doing everything from the command line and weren't very experienced with the Netinfo database structure. Still, it would have been easier to recover from our mistakes if the configuration information were stored in some plain text or XML format. Now that things are up and running, Netinfo hasn't been a problem. I think that using Netinfo to share user account information has been fairly easy, easier than setting up LDAP, although that's probably changed since Panther has been released.
One other note is that encrypted passwords are visible in our root Netinfo database to the outside world. This has changed since Panther, or so I understand.
Good question. All I know is that earlier this summer, a GUI frontend for DarwinPorts appeared in a Panther build, only to be withdrawn later. I'd like to know what happened too.
Also, what's going on with the metapkg alliance? Sounded promising, but doesn't look like there's much activity. Looks like Fink is the de facto packaging standard on OS X for now.
The referenced article is dated October 2002. Is this a mistake, or is this old news?
Anyhow, going to the Blue Gene web page, there is a document dated Nov 2002, an overview of BlueGene/L. An excerpt:
The approach we have adopted is to split the operating system functionality between compute and I/O nodes...
The compute node operating system, also called the BlueGene/L compute node kernel, is a simple, lightweight, single-user operating system that supports execution of a single dual-threaded application compute process...
I/O nodes are expected to run the Linux operating system, supporting the execution of multiple processes. Only system software executes on the I/O nodes, no application code.
On a related note, it appears no longer possible to reclaim the function keys by using the old "reboot into OS9 and reset the keyboard preferences" trick. Apple seems determined that I use my function keys for such uncommon operations as adjusting my screen brightness.
I know I'm in the minority, but I map lots of stuff to my function keys in Emacs -- and I try to have the same function key map whether I'm using Linux, Windows, or OS X. Now that's no longer possible. I'd give up Expose to get my function keys back.
Proteins are biological polymers that are produced in living cells; they are composed of amino acids whose sequence is translated from DNA. The reason why the genome is of such great interest is that proteins provide the "molecular machinery" of the cell, to put things crudely; the genome provides a blueprint on how to assemble proteins, and the diversity of proteins gives rise to much of the cellular functionality essential for life.
Determining the 3D structure of proteins is a very hard but essential part of learning how they work. Unfortunately, knowing the sequence of a protein (which you can derive from DNA) only gives hints about the 3D structure. There are a number of large computational projects such as Folding@Home and Blue Gene that are devoted to predicting protein folding from a 1D sequence of amino acids to a 3D structure.
X-ray crystallography is the traditional way of determining the structure of proteins; you basically analyze the diffraction pattern of X-rays from a crystal of the protein of interest.
Now to your question: a multispan transmembrane protein is a protein that typically sits in the cell membrane that encloses the cell (alternatively, there are other internal membranes as well). Most of these proteins pass through the membrane several times, back and forth. These proteins are very important because they are involved in cell signalling and transport of substances into and out of the cell; ion channels are a prime example of transmembrane proteins. But transmembrane proteins are also notoriously difficult to study and crystallize because they do not solubilize without detergents, and are challenging to reconstitute in their native form.
If you look in the Protein Data Bank, there are lots of proteins that have been crystallized; but only a very small portion of them are transmembrane. This year's Nobel prize in part recognizes advances in studying the structure and function of these important proteins.
The committe has been burned a few times in the past
One of the early examples of this is the award to Johannes Fibiger of the 1926 Nobel Prize for medicine. Fibiger had purportedly demonstrated that worms caused cancer in laboratory rats. Other researchers were unable to reproduce his results.
Probably as a result of this gaffe, American pathologist Peyton Rous had to wait 56 years before receiving the 1966 Nobel Prize for medicine, for his discovery of tumor inducing viruses.
According to a footnote in the article, "An education outreach program has been established based on 3-D animations of anthropomorphic figures called NanoKids. See: http://nanokids.rice.edu."
I guess these color figures better illustrate what the authors had in mind.
Nice analysis, thanks. It's too bad the build info isn't already available from the command line.
Perhaps this means that the withdrawn 10.2.8 release last week already had the OpenSSL update, but Apple couldn't report it until the OpenSSL group publicly announced it.
Thanks for the post. Apparently, the web page to look at is here.
I'm not sure why the OpenSSL version number wasn't modified. As it stands, machines upgraded to 10.2.8 last week do not have the patch, but machines upgraded this week will have it. And since the OpenSSL version number is the same, if you type "openssl version" in the terminal, you get the same result whether you're using 10.2.6 (old), 10.2.8 (old) or 10.2.8 (new).
At least "ssh -v" indicates that OpenSSH has been patched in 10.2.8 (both the old and new versions).
I don't see anything about OpenSSL in that link, only OpenSSH. The OpenSSL vulnerability was reported several days after that post.
Speaking of OpenSSH, it's possible that the OpenSSL vulnerability may render OpenSSH vulnerable. Apparently, the newer versions of OpenSSH do not use OpenSSL for signature validation; however, Apple uses a somewhat older version of OpenSSH (3.4p1).
It's a little unclear whether the new problems in OpenSSL have been patched. According to the CERT page, Apple is reporting the vulnerability as fixed in 10.2.8. On the other hand, I have a 10.2.8 machine that still indicates OpenSSL version 0.9.6i, which is supposedly vulnerable.
Again, on a side note, I wish Apple would allow security updates to be installed independently of the main bulk upgrade.
Isn't there a chicken-or-egg issue here? If I were writing commercial software for the Mac, I wouldn't assign any functionality to function keys because users who don't do the triple-reboot maneuver simply won't have them conveniently available under OS X. Of course, they could use the Fn modifier, but then a function key becomes an awkward two-handed maneuver.
As for using function keys, I'm usually a vim user myself, but I sometimes use Emacs because it has some fabulous programming modes that I can't do without. I map the function keys to things like setting and clearing breakpoints, spawning a shell, etc. I could assign these functions to "lower-keyboard" Ctrl or Meta-key combinations, but Emacs is already keymapped to saturation. At least the function keys start out pristine.
I've seen Doublecommand and uControl, as another poster suggested, but I've avoided them in the past. Like other kernel extensions, they have a history of causing kernel panics with OS X upgrades.
Perhaps things are better now, but I'd much rather have Apple give back the function key configuration option that we already had in OS 9.
Fortunately, the OpenSSH bug appears to be difficult to exploit. But I agree that Apple needs to unbundle security updates from all others.
I also don't like the way Apple hides security problems until they have a patch. They were one of the few Unix vendors not to issue a statement in response to the CERT advisory. I understand that a patch might need to go through QC, but Apple should still let their users know about the problem so that they can restrict services until the patch is available.
I didn't really have any serious problems with 10.2.8. The only odd thing was that I had to reboot one extra time because the screensaver kicked in and somehow hanged the machine. After that I repaired permissions (don't know why this helps, but it does) and everything was fine.
Except the function keys. For some reason, Apple insists on using the function keys for things like changing sound volume and turning up and down the screen brightness. I prefer being able to remap the function keys for applications, especially Emacs. The only way I know of to reclaim the function keys is to reboot into OS 9, toggle the boxes in the keyboard control panel, and then reboot into OS X. However, every software update of OS X forces me to repeat this process. That means three reboots on average for each software update (four this time, due to the problem above), which is pretty irritating given how slow my PowerBook starts up.
What I find curious is that although this is a longstanding issue, relatively few Mac users complain about it. I suppose it has something to do with most Mac users being accustomed to a mouse-centric experience rather than a keyboard-centric user interface. Or do people really need to change their screen brightness that often?
Slashdot Mirror
The link is on Google's main page. Look under the search box, it says today "Want a job that's out of this world?"
I don't think gmail is a joke. It looks like Google's real April Fool's joke is here:
http://www.google.com/jobs/lunar_job.html
It's always been hard to see distant planets because they don't emit light. Hubble can see distant galaxies because they contain lots of luminous objects.
Someone feel free to correct me if I'm wrong, but it doesn't seem that XGrid offers anywhere near the flexibility and power of Sun Grid Engine, which is open source and currently supports OS X among many other Unix platforms. SGE allows for tight integration of MPI processes, works for heterogeneous and distributed clusters, has lots of configurable options for remote monitoring, prioritization, and submission of jobs, and works well from the shell. The main selling point of XGrid appears to be the ease of configuration, which makes a difference particularly if your cluster of machines is "fluid" rather than fixed.
This class is in the Stanford CS193 series, which consists of lighter workload classes often taken by students outside the CS department, and geared for specific technologies. I don't believe CS majors get much credit for taking these classes. The core CS classes at Stanford are much less focused on details of specific languages (although Java is featured in the main intro OOP class; I guess you have to pick something).
.NET, Microsoft MFC, and Java, so a class in Cocoa is not too surprising.
Stanford has offered classes in
As another poster mentioned, Ctrl-V in vim has allowed rectangular selection for some time. In emacs, cua-mode allows rectangular selection as well, and it's easy to use.
I've heard nice things about BBEdit, but if you use several platforms regularly (not just mac), I find a cross-platform editor is preferable. Besides, it's unlikely that BBEdit will ever be able to support such specialty needs as a mode for the IDL language.
We run a lab with 5 XServes running OS X Server (10.2). On the one hand, Netinfo has been a convenient way of sharing user accounts across the XServes. We store user accounts, NFS exports, and DHCP configuration information in the head node and it seems to work ok. DNS stuff is configured with traditional BIND text files.
On the other hand, twice during configuration we managed to corrupt the root Netinfo database on the head node. Once we were able to recover from a backup; the other time we corrupted it so badly that we could no longer log into the machine and had to reboot into single user mode and do some nontrivial recovery.
Granted, we ran into trouble mainly because we were doing everything from the command line and weren't very experienced with the Netinfo database structure. Still, it would have been easier to recover from our mistakes if the configuration information were stored in some plain text or XML format. Now that things are up and running, Netinfo hasn't been a problem. I think that using Netinfo to share user account information has been fairly easy, easier than setting up LDAP, although that's probably changed since Panther has been released.
One other note is that encrypted passwords are visible in our root Netinfo database to the outside world. This has changed since Panther, or so I understand.
Good question. All I know is that earlier this summer, a GUI frontend for DarwinPorts appeared in a Panther build, only to be withdrawn later. I'd like to know what happened too.
Also, what's going on with the metapkg alliance? Sounded promising, but doesn't look like there's much activity. Looks like Fink is the de facto packaging standard on OS X for now.
The referenced article is dated October 2002. Is this a mistake, or is this old news?
Anyhow, going to the Blue Gene web page, there is a document dated Nov 2002, an overview of BlueGene/L. An excerpt:
The approach we have adopted is to split the operating system functionality between compute and I/O nodes...
The compute node operating system, also called the BlueGene/L compute node kernel, is a simple, lightweight, single-user operating system that supports execution of a single dual-threaded application compute process...
I/O nodes are expected to run the Linux operating system, supporting the execution of multiple processes. Only system software executes on the I/O nodes, no application code.
On a related note, it appears no longer possible to reclaim the function keys by using the old "reboot into OS9 and reset the keyboard preferences" trick. Apple seems determined that I use my function keys for such uncommon operations as adjusting my screen brightness.
I know I'm in the minority, but I map lots of stuff to my function keys in Emacs -- and I try to have the same function key map whether I'm using Linux, Windows, or OS X. Now that's no longer possible. I'd give up Expose to get my function keys back.
"Pointy-Haired Boss", from the Dilbert character.
Proteins are biological polymers that are produced in living cells; they are composed of amino acids whose sequence is translated from DNA. The reason why the genome is of such great interest is that proteins provide the "molecular machinery" of the cell, to put things crudely; the genome provides a blueprint on how to assemble proteins, and the diversity of proteins gives rise to much of the cellular functionality essential for life.
Determining the 3D structure of proteins is a very hard but essential part of learning how they work. Unfortunately, knowing the sequence of a protein (which you can derive from DNA) only gives hints about the 3D structure. There are a number of large computational projects such as Folding@Home and Blue Gene that are devoted to predicting protein folding from a 1D sequence of amino acids to a 3D structure.
X-ray crystallography is the traditional way of determining the structure of proteins; you basically analyze the diffraction pattern of X-rays from a crystal of the protein of interest.
Now to your question: a multispan transmembrane protein is a protein that typically sits in the cell membrane that encloses the cell (alternatively, there are other internal membranes as well). Most of these proteins pass through the membrane several times, back and forth. These proteins are very important because they are involved in cell signalling and transport of substances into and out of the cell; ion channels are a prime example of transmembrane proteins. But transmembrane proteins are also notoriously difficult to study and crystallize because they do not solubilize without detergents, and are challenging to reconstitute in their native form.
If you look in the Protein Data Bank, there are lots of proteins that have been crystallized; but only a very small portion of them are transmembrane. This year's Nobel prize in part recognizes advances in studying the structure and function of these important proteins.
The committe has been burned a few times in the past
One of the early examples of this is the award to Johannes Fibiger of the 1926 Nobel Prize for medicine. Fibiger had purportedly demonstrated that worms caused cancer in laboratory rats. Other researchers were unable to reproduce his results.
Probably as a result of this gaffe, American pathologist Peyton Rous had to wait 56 years before receiving the 1966 Nobel Prize for medicine, for his discovery of tumor inducing viruses.
Better late than never.
According to a footnote in the article, "An education outreach program has been established based on 3-D animations of anthropomorphic figures called NanoKids. See: http://nanokids.rice.edu."
I guess these color figures better illustrate what the authors had in mind.
Nice analysis, thanks. It's too bad the build info isn't already available from the command line.
Perhaps this means that the withdrawn 10.2.8 release last week already had the OpenSSL update, but Apple couldn't report it until the OpenSSL group publicly announced it.
Thanks for the post. Apparently, the web page to look at is here.
I'm not sure why the OpenSSL version number wasn't modified. As it stands, machines upgraded to 10.2.8 last week do not have the patch, but machines upgraded this week will have it. And since the OpenSSL version number is the same, if you type "openssl version" in the terminal, you get the same result whether you're using 10.2.6 (old), 10.2.8 (old) or 10.2.8 (new).
At least "ssh -v" indicates that OpenSSH has been patched in 10.2.8 (both the old and new versions).
I don't see anything about OpenSSL in that link, only OpenSSH. The OpenSSL vulnerability was reported several days after that post.
Speaking of OpenSSH, it's possible that the OpenSSL vulnerability may render OpenSSH vulnerable. Apparently, the newer versions of OpenSSH do not use OpenSSL for signature validation; however, Apple uses a somewhat older version of OpenSSH (3.4p1).
It's a little unclear whether the new problems in OpenSSL have been patched. According to the CERT page, Apple is reporting the vulnerability as fixed in 10.2.8. On the other hand, I have a 10.2.8 machine that still indicates OpenSSL version 0.9.6i, which is supposedly vulnerable.
Again, on a side note, I wish Apple would allow security updates to be installed independently of the main bulk upgrade.
What versions of OpenSSH would be vulnerable?
Apple OS X, for instance, still uses OpenSSH 3.4p1 (even after the recently withdrawn 10.2.8 update).
Perhaps they've filed for patent, as they've done before.
Isn't there a chicken-or-egg issue here? If I were writing commercial software for the Mac, I wouldn't assign any functionality to function keys because users who don't do the triple-reboot maneuver simply won't have them conveniently available under OS X. Of course, they could use the Fn modifier, but then a function key becomes an awkward two-handed maneuver.
As for using function keys, I'm usually a vim user myself, but I sometimes use Emacs because it has some fabulous programming modes that I can't do without. I map the function keys to things like setting and clearing breakpoints, spawning a shell, etc. I could assign these functions to "lower-keyboard" Ctrl or Meta-key combinations, but Emacs is already keymapped to saturation. At least the function keys start out pristine.
I've seen Doublecommand and uControl, as another poster suggested, but I've avoided them in the past. Like other kernel extensions, they have a history of causing kernel panics with OS X upgrades.
Perhaps things are better now, but I'd much rather have Apple give back the function key configuration option that we already had in OS 9.
Fortunately, the OpenSSH bug appears to be difficult to exploit. But I agree that Apple needs to unbundle security updates from all others.
I also don't like the way Apple hides security problems until they have a patch. They were one of the few Unix vendors not to issue a statement in response to the CERT advisory. I understand that a patch might need to go through QC, but Apple should still let their users know about the problem so that they can restrict services until the patch is available.
I didn't really have any serious problems with 10.2.8. The only odd thing was that I had to reboot one extra time because the screensaver kicked in and somehow hanged the machine. After that I repaired permissions (don't know why this helps, but it does) and everything was fine.
Except the function keys. For some reason, Apple insists on using the function keys for things like changing sound volume and turning up and down the screen brightness. I prefer being able to remap the function keys for applications, especially Emacs. The only way I know of to reclaim the function keys is to reboot into OS 9, toggle the boxes in the keyboard control panel, and then reboot into OS X. However, every software update of OS X forces me to repeat this process. That means three reboots on average for each software update (four this time, due to the problem above), which is pretty irritating given how slow my PowerBook starts up.
What I find curious is that although this is a longstanding issue, relatively few Mac users complain about it. I suppose it has something to do with most Mac users being accustomed to a mouse-centric experience rather than a keyboard-centric user interface. Or do people really need to change their screen brightness that often?
Ok, thanks for the clarification. The point release cycle just seemed somewhat shorter than usual (less than two months since 2.3).