I just received a call from an "IT recruiter" (I don't remember who, and it doesn't matter who, really), who started off the conversation by breathlessly exclaiming:

"I need to speak to someone about a network problem."

Um, who is this?

"Is this the IT department? I need to speak to the IT manager."

Um, that would be me. Who are you trying to reach? Who are you?

"I'm so-and-so, this is the number that they forwarded me to."

Um, there is no "they", we have an auto-attendant. Who is this again?

"I got your number from J. Random Otherperson."
I don't know them, but OK. Who are you?

"I'm so-and-so, and I'm with an IT recruiting firm, and I wanted to know if blah blah you had any projects blah blah" (Yeah, I figured this out by now, but I wanted to let it play out.)

Hi, well, why did you give me this whole story instead of just coming out and saying it? I don't like being told stories to. To tell you the truth, we're not inclined to want to work with people who lie to us. I certainly don't like being told a whole cock-and-bull story to get my attention. Thank you very much. Good-bye. <click>

I certainly would have listened and been polite and told the recruiter at the outset that no, we're not interested in talking to recruiters right now (and I know it's a very tough market for them, I really do not belittle their pain) but sleazy sales tactics in a field where success is defined a whole lot by trust just doesn't seem like a good plan--I suppose if you're using the "spammer" mentality of "try 100,000 and if 0.01% gets through, that's 10 sales" it might work, if you only had a short-term goal. But the good recruiters I've dealt with (and I've dealt with quite a few) spent time to cultivate a relationship of trust with clients, both on the buy and sell sides (i.e. employers and potential employees).

I hope it's not the same way in every sales arena.

The whole climate change stuff has really split the physics community. Behold the following letter I just received.

(I add my own commentary at the bottom.)

Dear fellow member of the American Physical Society:

This is a matter of great importance to the integrity of the Society. It is being sent to a random fraction of the membership, so we hope you will pass it on.

By now everyone has heard of what has come to be known as ClimateGate, which was and is an international scientific fraud, the worst any of us have seen in our cumulative 223 years of APS membership. For those who have missed the news we recommend the excellent summary article by Richard Lindzen in the November 30 edition of the Wall Street journal, entitled "The Climate Science isn't Settled," for a balanced account of the situation. It was written by a scientist of unquestioned authority and integrity. A copy can be found among the items at http://tinyurl.com/lg266u, and a visit to http://www.ClimateDepot.com can fill in the details of the scandal, while adding spice.

What has this to do with APS? In 2007 the APS Council adopted a Statement on global warming (also reproduced at the tinyurl site mentioned above) that was based largely on the scientific work that is now revealed to have been corrupted. (The principals in this escapade have not denied what they did, but have sought to dismiss it by saying that it is normal practice among scientists. You know and we know that that is simply untrue. Physicists are not expected to cheat.)

We have asked the APS management to put the 2007 Statement on ice until the extent to which it is tainted can be determined, but that has not been done. We have also asked that the membership be consulted on this point, but that too has not been done.

None of us would use corrupted science in our own work, nor would we sign off on a thesis by a student who did so. This is not only a matter of science, it is a matter of integrity, and the integrity of the APS is now at stake. That is why we are taking the unusual step of communicating directly with at least a fraction of the membership.

If you believe that the APS should withdraw a Policy Statement that is based on admittedly corrupted science, and should then undertake to clarify the real state of the art in the best tradition of a learned society, please send a note to the incoming President of the APS ccallan@princeton.edu, with the single word YES in the subject line. That will make it easier for him to count.

Bob Austin, Professor of Physics, Princeton
Hal Lewis, emeritus Professor of Physics, University of California, Santa Barbara
Will Happer, Professor of Physics, Princeton
Larry Gould, Professor of Physics, Hartford
Roger Cohen, former Manager, Strategic Planning, ExxonMobil

By the way, the folks signing this, they're not nobodies--they're major players in the world of physics, and widely respected.

The whole climate change problem is FAR from over, the major scientific societies are at each others' throats over this whole mess.

This whole tempest is over what constitutes legitimate "massaging" of data, and whether or not the climate scientists whose email was released did so. (There are many things one does with the raw data to normalize it in order to make sure you are comparing apples to apples, etc.) I have not spent enough time looking into what exactly these scientists did to their data, but none of them (as has been mentioned) is denying what they did.

One of the major tenets of scientific research is to be skeptical about all theories and data, requiring falsifiability for theories and independent repeatability for experiments. Unfortunately, climatology does not readily admit to either one--we can't well set up a controlled environment comparable to Earth, nor can we repeat long-term measurements. (Make no bones about it, too: there is much hemming and hawing about the validity and accuracy of the raw data, which contributes in large part to the statistical massaging that MUST be done in order for the data to be sensibly used in any climate model!) There are clear indications that something is going on--recent data on polar ice sheets are indisputable, but like everything else in science, without a model, one cannot know why the ice sheets are melting.

Unlike the other instances of scientific misconduct coming out recently (think the Bell Labs debacle of a few years back), this particular instance *begs* one to ask: cui bono? If there's something wrong going on, why is it going on? Who stands to benefit from all this?

Practically speaking, all this doesn't mean we shouldn't personally still reduce consumption and output of hydrocarbons, nor should we renege on our commitment to find alternative energy sources (without energy growth, our economy will come crashing to a halt, and the whole "Peak <X>" problem [for <X> in "natural gas","oil","water"] is a statement of mathematics, as indisputable as any other part of math--and the diminishing ability to produce more energy is certainly an issue) However, whether one should support, for example, "cap&trade" for emissions, or other new public policies that are based on science that is now cast into doubt, and come with extremely wide-ranging consequences, is something that you should very much reconsider. If anthropogenic climate change really is occurring, we do need to take active steps; if not, or if the effects are of equivalent order of magnitude to other naturally occurring events, one can still engage in them, but certainly you should not justify these policies with cries about impending climate doom--especially when sooner than climate doom we are facing major energy growth problems. ]]> tag:www.jbaltz.com,2009:/weblog//4.87 2009-11-20T15:52:02Z 2009-11-20T16:36:48Z

Heads up 1&1 users:

If you use IMAP to access your email at 1and1.com, you'll need to subscribe to a new folder called "Sent Items" to replace your "Sent" folder, and a new "Spam" folder in addition to a "Junk" folder (if you had it).

If you use Thunderbird, you should get the FolderFlags extension and use it to set the flag on your "Sent Messages" folder to "sent"

You may also need to check to see if you're subscribed to the new folders (if it doesn't appear automatically for you) and you may also want to set the 'Junk' flag on the 'Spam' folder. You'll also want to make sure that you configure Thunderbird to save your messages automatically in the new folder under Tools | Account settings:

If you use Outlook, you have to make substantially similar changes...

Of course, the email from 1&1 doesn't mention anything (read on)...and don't get me started about the whole "2.0" nomenclature (tagging along with "Web 2.0 Expo" that I just attended...that's a subject for a Random Rant)]]> Dear 1&1 Customer,

Your account and all mailboxes in that account, can now be used with 1&1 WebMail 2.0.

When you and your e-mail users login, you will not only notice the new layout, but also new features that are available to you.

WebMail 2.0 is included in your package for no additional cost.

The login link remains the same and, if you are using WebMail, you can continue to access it from any browser at: http://webmail.1and1.com

For more information about 1&1 WebMail 2.0 please visit our FAQ at: http://faq.1and1.com/search/go.php?t=n49907

We hope you will enjoy using the new WebMail as your primary e-mail program or in addition to your local e-mail client.

Sincerely,

Your 1&1 Internet Team
1&1 Internet Inc.
http://1and1.com]]> tag:www.jbaltz.com,2009:/weblog//4.86 2009-08-13T12:00:00Z 2009-08-20T17:10:10Z

blogger blogs on blogging, blog at 11

Jerry B. Altzman www.jbaltz.com Facebook and the "microblogging" I do on Twitter--not that in either case I have a great following (Although my 100+ "friends" on FB are an order of magnitude greater than the number of non-spam-followers on Twitter, which in turn is probably an order of magnitude greater than the number of readers of my blog...) I suppose part of the issue stems from the relative ease of publishing to either FB or Twitter (both have SMS update ability, so I can do it from out and about with my phone, and both have a bevy of browser plug-ins that make it possible to simply click on the status bar of my browser and say something, anything!), part of it stems definitely stems from the notion that a formal blog should Say Something Of Import™, which is somewhat at odds with the whole "world in 140 characters" of SMS--the stock in trade of Twitter.

(Of course, there is a facebook/Moveable Type connector that I just found now while writing this...and the one that MT puts at the bottom of my page, so we'll see how that works out. Certainly the ability to blog from FB into MT is nice, but even Facebook has a limitation of how many characters a status update can have makes that particular blogging channel suboptimal if the goal is to Say Something Of Import™) UPDATE 20 August 2009 Of course, there's nothing new under the sun: Jeff Atwood said it a while ago.]]> tag:www.jbaltz.com,2009:/weblog//4.85 2009-08-12T19:14:28Z 2009-08-12T19:27:19Z

We found what looks like a shark's head at Plum Beach

Jerry B. Altzman www.jbaltz.com went back to the same urban beach we visited just over two years ago. While we saw the standard hermit crabs on the beach, people catching fish, windsurfing, and the ubiquitous litter (come on, fellow city-dwellers! you can do better than this!), we also got to see something unusual: what looked to be a shark's head separated from the rest of its body:

Somewhat more closeup, this is what my Palm Treo could make out:

Certainly not your usual fare!

]]> tag:www.jbaltz.com,2009:/weblog//4.84 2009-07-23T02:47:55Z 2009-07-23T03:14:37Z

(Boy it's been a while.) So a few weeks back I had the pleasure of migrating a cilent's website from a shared web host + shared database host to a single VPS (virtual private server) where both the webserver and...

Jerry B. Altzman www.jbaltz.com So a few weeks back I had the pleasure of migrating a cilent's website from a shared web host + shared database host to a single VPS (virtual private server) where both the webserver and database server were on the same machine.

This consolidation is actually counter to the standard multi-tier architecture:

          +----------------------+
          |      webserver       |
          |                      |
          +----------------------+

          +----------------------+
          |       database       |
          |        server        |
          +----------------------+

In this case, we performed the consolidation because the shared database server that was available was horribly overburdened, and also the current application wasn't so much "architected" as much as "pieced together à la Frankenstein's monster" and wasn't uniformly careful with cleaning up its database connections, or even using database pooling (for .NET), so at times we'd see up to 4000 (!) separate connections attempting to be made to the database server.

While we were having awful problems with the database server, the technical staff at the hosting provider in question were helpful and responsive, even if they couldn't help us narrow down all the problems--some of which were caused, no doubt, by the client's own applications.

To tie in to the title, one of the big hangups came when we attempted to restore a MS SQL server backup of the original databases onto the local installation. One of the problems is that SQL server has a parallel notion of user IDs, namespaces (for tables and tablespaces and whatnot) and schemata (schemas), with some apparent overlap.

DISCLAIMER: I am not a SQL server database administrator (DBA) by any definition except by dint of having to manage a database on a single small server. Oracle, MySQL, PostgreSQL, these I can deal with, but SQL server? Not really.

So after restoring the databases, I found some standard permissions and ownership errors. And trying to just create a database user using the standard GUI method didn't work. (It never does. Why would anyone be surprised with this?). I found, however, that by dissecting the SQL that the create-a-user script uses, and modifying it ever so slightly, I could get things to work just my way. Perhaps this isn't the Microsoft way of doing things, and anyone who wants to disabuse me of my bad methodology should kindly step up and do so, as I am willing to learn.

The magic (line numbers inserted by me, actual databases and usernames modified, of course)

1  USE [master]
2  GO
3  CREATE LOGIN [loginname] WITH PASSWORD=N'password', DEFAULT_DATABASE=[defaultdb], CHECK_EXPIRATION=OFF, CHECK_POLICY=OFF
4  GO
5  USE [defaultdb]
6  GO
7  ALTER USER [loginname] WITH  LOGIN=[loginname]
8  GO
9  USE [defaultdb]
10 GO
11 ALTER USER [loginname] WITH DEFAULT_SCHEMA=[schemaname]
12 GO

The key was that the loginname and the schemaname, even though they are the same text string, are totally independent, and needed to be assigned. But in line 3, the standard CREATE LOGIN creates a disabled user, and that broke the ALTER USER statements in lines 7 and 11. (Or at least, that's what my memory has.)]]>

+-------------+       +---------------+
| web server  |       | database srvr |
| user 'root' | ----> |               |
+-------------+       +---------------+

mysql> show grants;
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Grants for root@webserverhost                                                                                                                                      |
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| GRANT ALL PRIVILEGES ON *.* TO 'root'@'webserver' IDENTIFIED BY PASSWORD '*you think i will put this here??!!' WITH GRANT OPTION                                   |
| GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, REFERENCES, INDEX, ALTER, CREATE TEMPORARY TABLES, LOCK TABLES ON `mysql`.* TO 'root'@'webserverhost'          |
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------+

mysql> grant all on *.* to root@'webserver' identified by 'xxxxxyyyy' with grant option
    -> ;
Query OK, 0 rows affected (0.00 sec)

mysql> flush privileges;
Query OK, 0 rows affected (0.00 sec)

[jbaltz@webhost] >mysql -u clover -pxxxxxyyyy -e 'show tables from newDataBase' -hDatabaseServer
+-----------------------+
| Tables_in_newDataBase |
+-----------------------+
| User                  |
+-----------------------+