But the problem isn't always on the recruiter's side. I'm my department's chief engineer and as such I see some of the resumes provided by applicants (submitted both directly and through a headhunter). In many cases the qualifications and experience cited bear little relevance to the skills required for the open position. I'll add that my shop doesn't necessarily treat the mismatch as an automatic disqualification, but it certainly doesn't help the applicant. The same was true of my PPOE (in the 1960s) where the system programming manager was a med school dropout and the lead system programmer was an architect.

As the saying goes, "it depends." I've written one resignation letter in my professional career - after 18 years as chief sysprog with an Enormous State University's computer center (and that was 32 years ago; I'm still with my second employer). In my case the reason for leaving was disgust with the university's administration - an opinion shared by the computer center director. I gave the director a heads-up almost a year before I left that I was looking for another position, and in my formal resignation promised to stay through a major installation to help it go smoothly. Since then I've twice been a manager (and both times escaped without serious injury) so I see the resignation letter from both sides. A few comments: * If the departure is amicable, say so. "I've been privileged to work with the highly professional staff at World Wide Widgets but have found new opportunities to develop and use new skills elsewhere". Volunteer to help plan the transition to backfill your position. Don't say that if it's not true, but in that case don't lie. * Note that if the departure is amicable (as was mine), ask your co-workers who think highly of your qualifications for permission to give their names and (perhaps personal) phone numbers to potential employers. Your employer will almost certainly refuse to disclose anything more than your dates of service, but that doesn't mean that you can't tell co-workers that you approve them going into detail. (No, this isn't part of the resignation letter but it's something to consider). * The "normal" expectation by an employer is a two-week notice that you plan to leave, but the employee manual, contract, or other document may require longer notice. If possible give a lead time long enough to make a smooth transition (assuming no need to say "I'm gone"), but consider whether that will trigger security procedures. Many companies have a policy that any employee with access to sensitive data will be terminated immediately upon receipt of a letter of resignation, then be given a payment equal to the salary they would have received between notice and departure (typically called "payment in lieu of notice"). * If you think it necessary to explain - either in the letter of resignation or in the exit interview with HR - problems that led to your resignation, BE POLITE and to the greatest extent possible, describe the problem and not the personalities involved. Even if your current employer refuses to disclose the details of your work a potential employer may hea

den2k88 wrote:

If you think 'goto' is evil, try writing an Assembly program without JMP. -- TNCaver

That's trivial: just code for a system that uses "B" [Branch] instead (e.g., S/360 et seq) or "TRA" [TRAnsfer] (IBM 704x/709x). Serious question: does any current-production assembler *not* use "JMP" (or some other mnemonic based on "jump")?

Back in the late 1960s - yes, I'm an old-timer - while working as the chief sysprog at a university I made major changes to the FORTRAN compiler used in IBM's 704x mainframe computers, adding features that were in later implementations of the language. Many of the features required that I keep tables within the compiler, and at various points I ran sanity checks to ensure that the tables were mutually consistent. When the sanity checks failed I failed the compile and issued the error message: ERROR 1164 - HOW IN THE HELL DID YOU GET HERE Only once in the several years that the enhanced compiler was in use did someone get the message, and it revealed a flaw in my code. (IIRC the bug was how I handled one-line DO loops.) On the subject of Easter Eggs, at one time long, long ago I maintained a list of Easter Egg sightings on the old WUARCHIVE server. Sadly, it's no longer in existence, having disappeared about five years ago (although much earlier I stopped maintaining the list of eggs because people stopped reporting them).

Mark_Wallace wrote:

It's the birthday of the "Space Telescope". I very much doubt that Edwin Hubble's birthday falls on the same day.

True (he was born on 20 November 1889) but to the general public the name "Hubble" is associated with the space telescope and not with the man. Last Saturday the Smithsonian's Air and Space Museum had a special day celebrating the Hubble, with several astronauts from Hubble missions talking about their experiences. I spent most of the day handing out NASA material on the Hubble to museum visitors; although in talking to the visitors I sometimes brought up Hubble-the-man I don't recall a single time that a visitor initiated a discussion of anything but Hubble-the-telescope. Fun fact: you can in part thank the Hubble for the CCD in your camera. Some of the early proponents of the HST wanted to use analog video to return images to Earth, but others, including NASA's first Chief of Astronomy (Dr. Nancy Grace Roman, in whose honor NASA named one of its four fellowships), held out for using the newly-invented CCD, and (after pushing its development to improve its size and sensitivity) they won - and as they say, the rest is history. For readers interested in a brief summary of Hubble-the-man and his contribution to cosmology, try http://www.pnas.org/site/classics/classics2.xhtml ("pnas" := "Proceedings of the National Academy of Sciences")

Member 10707677 wrote:

What happened to the standard business policy of not paying on an invoice unless accompanied by a purchase order. If the company cannot provide a valid purchase order number, the invoice gets returned to the vendor with a request for validation. If they cannot provide a valid matching purchase order number, it's a scam.

Part of the problem can be that for most large corporations, small purchases aren't handled through the purchase order system. In my experience with a relatively large R&D operation the department administrative aides have a corporate credit card in their name, with which they can pay for non-capital purchases that are below a certain amount...but what they order is (like a personal purchase) put on the charge card, and the AA keeps records of what was ordered. A delivery of unordered products stands out like the traditional sore thumb. An invoice that doesn't cite a matching PO won't be paid on arrival; if it's valid (e.g., a software maintenance renewal reminder) whoever is responsible for it will either arrange for the AA to put it on the corporate charge card, or initiate a purchase order. And if a vendor doesn't cooperate there are other approaches. My PPOE was a Very Large State University that back in the early 1970s had leased a high-speed (for that time frame) serial printer...but it didn't work. The vendor claimed that it had no obligation to give us a working product but that we still were obligated to pay for it. The vendor threatened to sue for payment, to which our purchasing department responded "we believe that we have more lawyers than you do." The vendor backed down and cancelled the lease at no cost to us other than elevated blood pressure.

mark merrens wrote:

I love the sheer, majestic beauty of these images. I know that they are quite often altered to enhance them; nevertheless that does not detract from what you see.

Don't forget that the vast majority of astronomical images - especially those from space telescopes - are not (and are not intended to be) representations of what the human eye would see, so by definition they have been "altered" by using colors visible to humans to represent other wavelengths. As you note, though, this in no way detracts from the beauty of the images. APOD also demonstrates that NASA has a sense of humor: look at the 1 April image for any year, but be prepared to explain to anyone nearby just why you're laughing. My personal favorite is the image posted on 1 April 2005 [^] which presents proof that there is water on Mars. An index by date to the images is at [^]

>back when I did do it, the cost of maintaining code far exceeded the >cost of developing it, and I considered a lack of meaningful comments >grounds for termination. I still do. +1 A friend of mine teaches programming for high-school students - and docks a student's grade if the programs aren't reasonably commented. Back in the early 1960s while in college I made some changes to a copy of Spacewar, and still have the source listings...but I have no idea today what those changes were. I hadn't learned about commenting (and in general neither had the other programmers who worked on the code). Now, as the senior engineer in my department at my POE I use those uncommented listings as examples of poor programming practice. (There are few things so horribly bad that they can't be used as a bad example.)

Eddy Vluggen wrote:

Dave Kreskowiak wrote:

We buy machines and they come with a Win8 license. The first we do is scrape 'em down to bare metal and slap it with the corporate Win7 load.

..and that's cheaper than a PC without an OS? I mean including the extra work you put in.

Actually...yes. Enterprise volume licenses for Windows are incremental on top of existing Windows licenses; they can't be used by themselves. Typical corporate purchases of new machines include an OEM license for the least expensive business-class Windows product offered by the manufacturer; once the machine arrives the hard disk is wiped and the corporate VL image is installed. Of course, for some hardware the manufacturer might not offer a Windows client. At my POE a few years ago we bought ~50 low-end servers to provide service quality monitoring across the network. The monitor tool ran on Windows 7 Enterprise, so we bought 50 copies of Vista Business, from which we cut out and saved the COA (to prove ownership of what Microsoft calls the "qualifying license") and threw the rest (including the disks) into the trash. That's more work (and more expensive) than would have been the case had we been able to buy an OEM Windows license as part of the hardware purchase.

Agreed about robocopy, but especially the version that ships with Windows 7 (or Vista if you must). Earlier versions don't understand daylight saving time. (see the /DST option) Other tools for me include REG, WMIC, kedit (the port of the mainframe editor xedit, not the linux-based tool) and (do'h!) the suite of commands built into CMD.EXE. Ditto for PowerShell, but see the comment below. And, of course, the Sysinternals suite from Mark Russinovich. I occasionally get a bit of ribbing for continuing to write .CMD scripts but (perhaps influenced by having started my IT career with character-mode interfaces - on punched cards, 50 years ago) my position is that the command line is simple, well-understood, and stable; if I need to do something that can be done via a .CMD script that's what I use; if it can't but still needs to be scripted I use PowerShell, or just write a new command-line tool.