What to do?

[crfriend]

I've never heard the "adage": "There's never enough time to do it right, but always enough time to do it over." We always used to say: "Do it right and do it once!"
Another was: "If a job's worth doing it's worth doing well." But that might have been in times gone by.

"There's never enough time to do it right, but there's always time to do it over" is a sarcastic observation describing what happens when a craftsman is forced to do something too quickly by an impatient superior who then finds out that something was overlooked (because of the accelerated schedule), blames the craftsman and tells him to go back and fix it. This culture is not universal by any means, but it's rampant around here. And it's not new: It was the root cause of the Apollo 1 fire that killed three people.

I like "Do it right and do it once!" and "If a job's worth doing, it's worth doing well." However, those seem to have slipped below the waves before I took up a professional mantle. I try, however, to live by those. I guess I'm a bit of a throwback.

[kingfish]

Toyota just recently settled a lawsuit involving bad firmware involved with the throttle. Remember the people whose cars accelerated out of control on them? Come to find out the code for the pair of "redundant" micro-controllers that watched what the accelerator pedal was doing hadn't been validated, nor were they set up to use independent data sources. They cut corners and people went screaming down the highway, burning out their brakes in a vain attempt at stopping as a result.

I also blame the likes of Microsoft and those that push the envelope on the principle that the best is the enemy of good enough. Those guys built their monopoly cashing in on the low expectation of reliability that most of the public had for the computer.

[Sinned]

At a previous job we had the Corporate motto "Right first time" among others and what amused us at the time was that this and the other mottos were put on pencils with a rubber ( eraser ) on the end. I also like the saying about providing every assistance short of actual help. Isn't the English language wonderful? I often wonder if there are similar subtleties in otherlanguages. Maybe it's a by-product of incorporating so many other languages into our own.

As a hard tasking Team Leader I always used to insist that any rework was done in the Developer's own time after work. Didn't increase the project's costs any because we were all salaried and very well paid. Did cause them ( and me, I was not exempt ) to think a bit more and to get it right first time. Ain't I horrible ( or was ).

[crfriend]

Isn't the English language wonderful?

Indeed it is, and one can have so much fun with it. Just today, I had two very hilarious moments that were entirely down to the vagaries of English if one wants to play it that way, and I've been trying to cook an excuse to publish them -- and this is it!

I had a delightful lunch today with a fellow systems administrator who I worked for a little more than a decade ago, and the connection was almost immediate. I thrive in the company of intelligent people, and over the course of a little more than an hour we'd touched on subjects as diverse as high-energy physics, rocket engines, explosives, and the usual panoply of computing topics.

During the time a tap at the bar "kicked" and managed to shower me with a certain amount of liquid (it turns out it was cider); humour being important, I feigned my best cartoon voice and uttered, "They got me!" Much laughter was had, including by the barkeep who immediately presented me with a pile of napkins to soak up the liquid. "Fortunately, it's a dry cider!" came from across the bar (a sense of play, too! This is good!). "Funny, it doesn't feel that way." (more laughter, and some commentary about satire ensued)

On the way home I figured I needed some more beer as my stock was running low and the Holiday is coming, so I stopped in at one of the places I frequent and picked up a 12-pack of Bass and a container of peanuts (an occasional treat I like almost as much as pistachios). "Would you like a bag for your nuts?" I cocked my head off to one side and look at the cashier with one baleful eye. "Pardon?" (laughter all the way 'round) "I need to be more careful with my wording!" (and a couple more chuckles)

Language is an eternal source of amusement for me, and one that I try to share with others when I can.

As a hard tasking Team Leader I always used to insist that any rework was done in the Developer's own time after work. Didn't increase the project's costs any because we were all salaried and very well paid. Did cause them ( and me, I was not exempt ) to think a bit more and to get it right first time. Ain't I horrible ( or was ).

Generally speaking, good people aren't put out by the expectation to "get it right the first time" -- so long as they're given the tools and the latitude to make it so. It's when you task people to do so, and then deny the resources needed that a disservice is being done.

[BobM]

Ah, yes. Get it right the first time. Back in the 60's I was the leading torpedoman on a Tench class diesel electric submarine (SS-483). Orders came down that we were to lay a practice mine field off the northern coast of Florida. O.K., that is one of the things that we were designed to do. We loaded 15 or so moored mines and set off to do our duty. Here is how it works: The torpedo tubes are 21" diameter and are, naturally, primarily designed for firing torpedoes. There are two devices that project into the top of the tube: a stop bolt that engages a stop on a torpedo--or mine--to prevent it from sliding too far forward, and a finger of bronze called the "tripping latch" that activates a steam torpedo when fired. Got all that? Great. We got down to Florida, about two miles of Jacksonville Beach and commenced to lay our mine field. There is (or was) a Navy occupational specialty called Mineman. We didn't have one of those, and the one who was supposed to come along on this evolution never showed up. None of us had ever layed mines before but how hard could it be? Now these mines were about 15 feet long and consisted of two parts: the forward 12 feet contained the business part, and the after 3 feet was a cast iron anchor that was supposed to separate, sink to the sea floor, and reel out a preset sea pressure controlled amount of mooring wire (hence, "moored mine"). Part of this assembly was a circular fin like arrangement called the balance ring. That ring kept everything copacetic until the moor was complete. Laying a mine field is a fast paced operation. There wasn't time between shots to even properly drain the tubes to the wrt tank. We just closed the outer door, vented pressure inboard, and opened the breach door. About 4,000 pounds of seawater poured into the bilges with each shot. We were using tubes 1 thru 4 in rotation and pushing one out every 2 minutes. We were kind of concerned that we were finding spare parts coming out with the water, but since the Weapons Officer didn't seem concerned neither was I. It turns out that, surprise! We should have been concerned. Remember the tripping latch? We didn't think to rotate it up out of the way since there was nothing on the mine for it to trip. What it did was tear off the balance rings. We found out later that most of the mines, being unbalanced while trying to set themselves simply tore loose and drifted up on Jacksonville Beach to the great consternation of the tourists. The Navy was not amused.

We had to rush to get this assignment done, didn't have the local expertise to do it, and were not provided such expertise by Squadron. The result was predictable.

[crfriend]

We had to rush to get this assignment done, didn't have the local expertise to do it, and were not provided such expertise by Squadron. The result was predictable.

Eeek! (If for nothing more than putting 4000 pounds of water into the "people tank" with each "shot".) I'm rather surprised that you were allowed to commence without the specialist on board, but since you were, and it was "fast-paced" the outcome was completely and entirely predictable.

Personally, even as a layman (and that has nothing whatsoever to do with laying mines) I have the common sense to know that if you've got hardware coming back into the cabin following a launch then you've got "problems" -- and I don't much care what the Weapons Officer has to say about it. If you've got hardware flushing back into the cabin with each shot, what's to say that something else didn't get caught up someplace else down-tube and kept a subsequent shot from going properly.... (Yes, I am partially using gunnery terms here, but lots of the basic principles remain -- if the muzzle is blocked and the breech-pressure gets too high...)

[Jack Williams]

I well remember the saga of Phillips compact cassette. Chucked onto the market for the customers to find out the faults, which were "fixed" from there. Never used them, but as a tech in those days, did I do a lot of work on those machines before they were finally vanquished by the (Phillips!) CDs! Always stuck to my reel-to-reels.
Strangely enough, with 50th anniversary of the little buggers there's a nostalgic revisiting of them, on the basis that they are, well, analogue after all, and analogue is definitely on the rebound. Read in a music publication recently that CD sales were down 15% but LP sales up 20%. Says it all.

My own opinion is: when it comes to audio "software" the ones to have are LPs and DVDs. The vinyl 12" 45s, instigated by HiFi enthusiasts and then boosted by DJs are the king. The 7" 45s sound bloody great too, and suit jukeboxes mightily.

[crfriend]

I well remember the saga of Phillips compact cassette. Chucked onto the market for the customers to find out the faults, which were "fixed" from there. Never used them, but as a tech in those days, did I do a lot of work on those machines before they were finally vanquished by the (Phillips!) CDs! Always stuck to my reel-to-reels.

I still have my 4-track (or plain "stereo") Akai reel-to-reel (with the big ones) and love it. It doesn't get much use today (the cats are way too interested with it when it's in motion) but I can clean it off and fire it up pretty much any time I want. It's sound quality is unsurpassed by anything other than broadband FM recording (which can go right down to DC) and it's a joy to watch in action.

Strangely enough, with 50th anniversary of the little buggers there's a nostalgic revisiting of them, on the basis that they are, well, analogue after all, and analogue is definitely on the rebound. Read in a music publication recently that CD sales were down 15% but LP sales up 20%. Says it all.

In all honesty, CDs trumped what was getting passed off as proper vinyl by the time that the late '80s rolled around. The sampling frequency guarantees that one will get the proper harmonics even if one might not be able to perceive a pure tone at near 20kHz, and if properly done the noise level is that what's ambient where the thing was recorded (that was a big problem early on with the technology as it'd capture the HVAC systems in studios).

I've resisted replacing much of my main-line hi-fi system over the years, but it may be necessary soon due to dust contamination of the physical bits (e.g. speaker-coils). However, the last time I fired it up in earnest it was accurate enough to not only tell me the difference between 44.1kHz uncompressed digital but also between three separate MP3 encoders -- and I could identify those by ear on multiple different musical selections. I have since abandoned MP3 and have started using FLAC; my ears thank me, even using mid-range computer speakers.

[Jack Williams]

Yeah, recording on vinyl from digital mastering is a bit naff. I guess the first stereo records, mastered with Scully lathes using all tube amplification were the ones. People are resurrecting that gear these days, and we are getting some really good recordings again.

As for going down to DC, that's not the problem (what do yer speakers make it to?) and the recording industry standard is 20hz to 20khz but it's the tops that are the hard part. A cassette at 1.75 IPS is gonna have a problem with that. A reel-to-reel at 7.5 IPS is not such a prob.
20 to 20 is no proplem for a good pickup, but what are one's ears like? Well, mine are good at last check for about 15khz, not bad for my age, but that's not the problem. It's the RISETIME! The steepness of the rise from zero to max on the recording. The attack. That is what you notice on a good recording. Does it sound mushy or sharp? Dont need 100kz ears to notice that, but ideally need 100kz recordings to percieve. I notice that on 12" 45s whether the official bandwidth says it or not. They manage to cut some good risetime on some of those vinyl records, but it's really chopped on digital.

[crfriend]

Yeah, recording on vinyl from digital mastering is a bit naff. I guess the first stereo records, mastered with Scully lathes using all tube amplification were the ones. People are resurrecting that gear these days, and we are getting some really good recordings again.

What about the days of pure acoustic recording? (Yes, there is a hint of sarcasm in there, but perhaps not as much as you might expect.)

As for going down to DC, that's not the problem (what do yer speakers make it to?)

Well, the speakers are good to DC but all that does is shove the cone out by a bit and then not move (this is how computer disk-drives position their heads by the way)

[...] and the recording industry standard is 20hz to 20khz but it's the tops that are the hard part. A cassette at 1.75 IPS is gonna have a problem with that. A reel-to-reel at 7.5 IPS is not such a prob.

However, at the same time, a moving tape at that speed can begin to lose the bottom end because the magnetic flux isn't changing quickly enough. It's a balancing act. 1 7/8 ips is a loser when it comes to the high-frequency end.

20 to 20 is no proplem for a good pickup, but what are one's ears like? Well, mine are good at last check for about 15khz, not bad for my age, but that's not the problem.

At last check, and I'm still astounded by it, mine are still good to about 17 or 18kHz (although I'm a few years younger if I recall, but given what I did to the poor things when I was a youngster...)

It's the RISETIME! The steepness of the rise from zero to max on the recording. The attack. That is what you notice on a good recording. Does it sound mushy or sharp? Dont need 100kz ears to notice that, but ideally need 100kz recordings to percieve.

A 100 kHz rise will happen in one sample interval at 44.1 kHz and will be so reflected unless it's a very sharp harmonic where it might be masked (e.g. a sharp square spike that returns to the original level within 2 or three sample-cycles), but those sorts of sounds do not usually occur naturally or with physical (i.e. non-synthesizer) instruments save for in the wild kingdom (e.g. bats). I'm not rubbishing your assertion by any means, but am trying to think it out.

I notice that on 12" 45s whether the official bandwidth says it or not. They manage to cut some good risetime on some of those vinyl records, but it's really chopped on digital.

I'm not willing to do the math at this time (being rather inebriated) but what would the tracking angle be for a high-amplitude pulse at 100 kHz present on a 45 RPM record at, say, 5 inches from the hub be? I'm thinking it'd look like a wall to the tracking-stylus (not that the cutting-stylus would be able to make the motion either).

[Jack Williams]

I think the cutting stylus would have a better show of cutting it than the replay one to reproduce it. But the 12" at 45RPM would have more of a slope for the stylus than the slower speed.
I basically flunked maths at school, but it is basic physics and the ears tell a good story.

[Jack Williams]

What indeed to do? I notice a lot of ahem smoking paraphinalia in that shot of the DJ.
When I showed him the shot he said: "Awsome". Excellent. I bet his records sound good in the club. Genious to have him out on the street in the entrance.
That was in Karangahape Road, Auckland New Zealand in the middle of the afternoon as I happened to be passing by on the way to the bus.
Sorry about the spelling but I flunked that as well besides getting top in English.

Above rave caused by some altered state of mind by afformentioned intellegence enhancing substances.
I'll think of what I really wanted to say, but it escapes me right now.
Seemed to have to do a lot of editing to get that somewhat right.

[crfriend]

Well, the Living Computer Museum in Seattle got a Christmas present from me today in the form of a new disk-device scanning routine that fully-generalizes any MASSBUS device that one cares to code in and present to the system -- in FIVE LINES of assembler and two tables.

Code: Select all

RPXCP1: MOVSI   T2,DODT         ;READ DRIVE-TYPE REGISTER
        PUSHJ   P,DODTI4
        LDB     T3,[POINT 9,T2,35]
        SETZB   T1,T4           ;END OF LIST CONSTANT, DRIVE INDEX
        CAMN    T1,MBDVID(T4)   ;END OF LIST?
        JRST    RPXCP3          ;YES - NO SUCH DRIVE TYPE.
        CAME    T3,MBDVID(T4)   ;DRIVE TYPE MATCH?
        AOJA    T4,.-3          ;NO - GET NEXT ENTRY

The first three lines are the original code and the last five mine, and I'm rather proud of it.

It's yet to be proved in operation, but in a diagnosis-run on paper, and then in simulation, it works as designed.

My word, getting the heave-ho from the last place I was working was one of the best things that's ever happened to me. Doing something like the above would have been unthinkable five months ago.

[arkie]

We used Emulex controllers (sc21/41?) with CDC 9762 and 9766 drives. We eventually got an RM05 with the washing machine sized Massbus controller.

What class of system was that on? VAX, pdp11, or something else? Nobody but DEC used MASSBUS as DEC owned the patents on some of the things, notably the connectors. (I don't care if they're ZIF (Zero Insertion Force) -- you can still bend the ruddy pins on 'em.)

I'm working on a PDP-10 system.

Now all of that data would fit on the smallest drive in my Clariion array.

And is possibly several orders of magnitude faster. Ain't technology grand?

We had a pair of PDP-11/70s and an 11/45. One 11/70 ran RSTS/e and the other ran RSX-11m+. We eventually got a MicroVax (3100?) and had them all DecNetted together.

Now I manage several EMC Clariions, two VMaxes, a VNX box, a couple of Celerras, two Centeras (yuck). I also have a furball of a SAN fabric with 65 or so Cisco switches.

Oh, I wear skirts on occasion and have even ventured out in public, but only in a very few, select, locations, like the thrift stores where I have purchased them. I also have several lovely polo dresses. So much for a fashion statement.

[Jack Williams]

Bamboosled again, but my computer literate friend Simon might be intrigued by that. I'll give him a look at it when he's here next.