Digital Cameras in Technical Communication

["F Cameron Sipston" <c_sipston -at- hotmail -dot- com>]

Digital cameras in Technical Communication

Some time ago, when I was actively evaluating potential jobs passed by
Internet, I was invited to interview for a contract which offered the chance 
to work in that sector of the craft best described as where proprietory 
operating instructions meet bespoke training materials.

Some criticisms of the recruitment process, based on this particular agent's 
assumptions about mobility of labour and socio-economic demography not 
translating effectively into contemporary economic reality, notwithstanding 
(the interview was to be at a roadside restaurant/motel simply not 
accessible without an automobile - not much use to out-of-town applicants 
reliant on trains and buses but capable of temporary relocation), the job 
was documenting machinery for use in commercial food packaging. And the 
documentation consisted of the training and operating materials for the 
machinery, presumably with temporary workers in mind.

Now, with a bit of goodwill and permissive schedules I could have blagged a 
lift to the venue but, personally, I was dissuaded by the mention of a 
digital camera for use in taking photographs of the machinery.

Now I'm a competent, but not expert, photographer with long, perhaps rusty, 
experience in monochrome film work, particularly the use of filters in 
enhancing detail through contrast, as well as experience in colour work and 
the obligatory grasps of concepts such as composition texture, film grain 
and how ambient lighting has traditionally called the shots WRT field-focus 
& shutter speed.

Yet the prospect of a digital camera struck doubt into my mind. And visions 
of being "left to it" with inadequate materials by a fast-talking agent, but 
being held accountable by the clients for a job badly done, haunted my 
nights.

Eventually, after assessing everything, I cancelled the interview. But I
wonder how the job might have turned out. Oh that I'd had an indemnity
policy at the time...

My experience with digital cameras

At this point I had no experience of digital cameras. Since then I've seen a 
few, results from more, and I've tried, unsuccessfully, to borrow my Dad's - 
he seems worried, notably because his own results have been inconsistent 
(which suggests, to me, that a second opinion would be valuable) and ,also, 
that I might somehow "break" it [1].

And I've also had a digital camera demonstrated to me by a friend, back in 
the country for PhD tuition but tenured overseas [2], the only difference 
between my friend's camera and domestically-available models being, I was 
assured, of cost compared to the effective RAM in the camera. How RAM 
available in the States compares to that here in the UK I don't know. Where 
my Dad's has 30-40 pictures at an acceptable screen resolution his offered 
about 400-600 for a comparable outlay.

The results? Sadly the overseas camera went missing from the university
halls, thought stolen. So I never got to see how those pictures of me turned 
out past a preview through the case-mounted LCD on the back.

Sadly because I was keen to see how they took being manipulated, refined, 
enhanced and suchlike on-screen as compared to prints digitised with a 
desktop scanner.

It was too late for the contract by this time. But certainly not too late 
for others.It was only later, when spending more time scrutinising previews 
on digital cameras, I noticed the effects, familiar to lap-top users, that 
both the physical screen angle and the crystal polarities had on the way the 
pictures previewed with this system looked at any given moment.

JPEG and digital cameras

It is, perhaps, both a boon and a limitation that the JPEG standard utilises 
different levels of compression. Those I've received through e-mail in the 
past which have been compressed at a comparatively high standard have, on 
older kit, displayed with dusts of anti-aliased pixels approximating tonal 
variation, but, not in any obviously filterable manner. Though when viewing 
the same images with a newer computer (but, notably, the same monitor), some 
time later, they rendered perfectly on screen.

And on opening my first images taken using a known digital camera,
compressed at JPEG level 6 with 3 scans, I noticed that, though they look 
impressively crisp at apprehensible zooms (50% for a 72ppi RGB image), when 
you get down to a pixel-level close-look (say, 100% at 72ppi or 25%-50%at 
300ppi), noise and distortions are very noticeable in places.

This seems to be most noticeable in areas of both high contrast and fine
detail - round the ears of a subject wearing a dark wedding suit standing in 
the sun-drenched doorway to a darkened church [3].

The effect is comparable, in some ways, to "camera shake". Indeed, this was 
what I first thought it was. But the distortion was localised. And classic 
camera shake affects the entire shot. Also, there was no comparable 
distortion noticeable at highly mobile, frequently gestural, extremities 
such as the hands (think of how quickly and how far you can move them 
compared to your ears - and the rest of the groom's face rendered just 
fine). As such the quality loss appears neither to result from camera shake 
nor from subject-movement (my second hypothesis).

Admittedly, movement is not something one would expect from the control
panel of industrial machinery one might be engaged to photograph. Fine
detail and marked contrast, particularly involving curves, however, are
often characteristics of such artefacts and the way we, as technical
designers, use friendly, round, buttons and high-contrast curved edgings as 
our stock-in-trade for functional delimitation in interface design. And very 
effectively for day-to-day operators.

Now, as a caveat, it's possible this experience simply reflects the limits 
of JPEG level 6 compression and the originals are photo-perfect. I do, 
however, doubt this. If the originals were photo-perfect then it seems to me 
there would be more, generalised, data-loss in the shots as a whole.

And, even if the originals are photo-perfect, it hasn't taken a massive
compression-step, assuming they were shot at a higher, home-printer
friendly, resolution of 150 or 300 dpi (but, remember, memory is a known
limiting factor with this camera), to irretrievably lose critical 
information.

Again this seems to be a scale thing. Blowing the precise contour of the
groom's ear up to 3x magnification at 2400dpi is of no importance to me,
personally, and nor will it be to most viewers of these particular photo's.

But the groom's ears are as large in real life as the largest of round
buttons on, say, a photocopying machine. And the distance which would be
required to photograph industrial machinery is comparable.

Even with a massive camera RAM, something a look round English photographic, 
electrical and computer shops still suggests is unlikely to be manifest 
cheaply for some time, the scope for error in digital photos for top-end 
composition by a print house running at, say, 1200 or 2400 dpi counsels 
caution.

And not only caution but time-consuming test runs and detailed negotiation 
with the print-house and, implicitly, added budgetary requirements. Things I 
got the impression the digital camera had been specified as a project tool 
to negate in the first place. And not something anybody involved seemed 
comfortable talking about when I asked directly (indeed, it was the way 
various assurances the camera was fine signified what was clearly a closed 
subject that prompted me both to cancel the interview and to get a better 
idea of the reality of digital photography).

Thus, while it's arguable that for some applications - such as inclusion in 
an Acrobat, or HTML, technical marketing document or workplace instruction - 
the digital camera would prove adequate, as this contract specified the 
deliverables would be paper-based and sent to a print-house it seems my 
bells of alarum were well-rung and, indeed, cancelling the interview was no 
foolish surrender at all. Just think about the shots of cars, bikes, boats, 
even computers, in the technical (hobby) press...the way you need to visit 
the showroom because the ad or review doesn't quite answer your questions on
what parameters that panel of indicators to the left of that prominent meter 
in the photo' relate to...

And, as such, having considered, assimiliated and now tested digital cameras 
from a basic competence with analog (chemical) photographic media, technical 
competence in market leading image processing applications and access to 
both a reasonable scanner and digital images from a known source, I can only 
suggest the scope for image degradation and data loss in digitally-generated 
photograph formats is something anyone considering working with digital 
cameras in technical photography for the first time should be-a-wary of.

And, as such, I can see drum scanners remaining with us quite some time yet.

Lighting and digital cameras

Back to the camera at the wedding. My experience as a subject [4] of the
same camera, later on, at the somewhat darker "evening do", without which no 
British wedding is complete [5], demonstrated nicely that the
multiple-scanning processes on which digital cameras work may require some 
following winds, and flashes configured around this over-sampling gave me 
borderline palpitation level shock and an irritable streak.

This was mainly the result of being bombarded, without warning, by intense, 
repeated, flashes of strong light in mid conversation (presumably in the 
interests of taking "candid" shots) and wasn't a reaction I could control. I 
do, now, wonder whether the red eye effect is minimised solely because of 
involuntary blinking rather than any bona fide application of advanced 
technology. The question of whether or not this might trigger a petit-mal 
fit in anyone prone to epilepsy is one I'll wait awhile [6] before asking.

But, surely, they wouldn't be on sale if they did...

Conclusion

As it is, my experience leads me to suggest that rookies out there,
particularly those without extensive photographic experience, and those with 
a background specialising in writing or DTP but under  corporate, agential 
or other financial pressure to branch out into the new, beckoning, wider 
world of fully integrated communication products by taking their own 
technical photographs, do spend some time looking into why many traditional 
applications of technical photography have used such bulky, costly, apparati 
as tripods, close-up lenses, composition beds, benches and tables, and 
complex flash-rigs to capture the required levels of detail in images on 
transparency before they jump into contracts which offer a digital camera 
not only as a working tool but also as a job perk or completion bonus.

Remember when you got that fantastic computer system that could do all kinds 
of previously complicated things at "the click of a mouse"? And look how 
many people still write here for help the first two years they're learning 
to use it...

So, for now, as far as digital image processing is concerned, whilst it may 
be true that 2400 dpi laser-setting boasts no discernible difference from 
analog printing, even to a typographic purist, and renders any kind of curve 
smooth and crisp, it seems digital cameras may well still have to play 
catch-up before we can all dash out to buy one tax-deductible as working 
capital and expect screen-resolution fidelity to translate to print-shop 
accuracy. Digital cameras are, however, just fine for weddings, barbecues, 
birthdays,
holidays, picnics, rallies, meets, a day at the races, beach parties &c.

And I'm sure they're well-suited to under-the-hood shots off the engine 
layout in the new-style Cadillacs recently launched into the UK market.

Ideal for paparazzi applications also I'd've thought. But, please, be 
careful with that flash, Eugene, when you're in a dark room at a party!