The Grand Illusion
Last year, because of my travel schedule, I finally replaced the old laptop I hadn’t used as a laptop for almost a decade with a new one. Since my travel and appearance schedule is irregular [to say the least], there are times I don’t use the laptop for days at a time, but for various reasons, I didn’t even open the laptop for the last three weeks of July. When I did, I was faced with “urgent warnings,” necessary updates, and priority downloads. Now, when I was using the laptop every few days, I didn’t quite grasp how much time was required to keep the device “updated.” I knew I was always updating the tower computer in my office, but never paid much attention to exactly how long it took, because I could just work on the writing computer [which is not connected to the internet] while all the updating and installing took place.
This time I kept track. The required updates for the laptop took almost four hours when I first signed on, and another two hours after I signed off. Add to that another hour after I signed back on after all the updating. Seven hours required in a little more than half a month less than a month. Oh… and the computer informed me that it has been required to make 16,060 update operations.
Now… so far as I can tell, with the exception of virus and security protection, which, interestingly enough, took almost no time at all, almost none of these updates measurably improve my computers’ operation and speed. In fact, it’s likely that each one marginally degrades their performance, and each one marginally pushes the software toward greater and greater problems because every change affects the operating system in some way or another. Yet, if I don’t upgrade, before long I can’t use other data, can’t open documents produced by “upgraded” systems, etc. So I, and every other user, am effectively being blackmailed to continually upgrade.
Why?
As I’ve noted before, and more than once, a great majority of “upgrades” are nothing of the sort, but consist of either patches or fixes or represent the grafting of more and more features onto existing programs, making them harder and harder to use for all but the geeks who never found a new feature they didn’t love and adding more time to my workload to learn the impact of the changes. This all reminds me of “the Red Queen’s race” in Alice through the Looking Glass, where the Red Queen announces to Alice that running as fast as she can is only sufficient to stay in the same place and that to get anywhere she must run even faster.
What ever happened to the idea of “getting it right” in the first place? Is the tech marketplace so fixed on being first that it’s required to come out with a new product before it’s ready… and then dribble out the fixes until it’s time to issue another new product that’s not ready?
To paraphrase Billy Joel, “if this is moving up, then I want to move out” … except, like everyone else who relies on computers… I can’t.
I think some of that’s a little unfair, if I’m honest.
Your computer’s operating system isn’t created for that one specific set of components, peripherals and software packages, it has potentially billions of combinations, if not more. Installing one piece of software might fix one thing, but it could cause an error elsewhere due to a conflict or a faulty piece of hardware. That same update may introduce a security hole that wasn’t seen. When you consider the amount of code and number of people working on a project, combined with the number of potential scenarios from past, present and future, there’s a lot of room for error. Updates have to be released to factor in situations that weren’t foreseen or dealt with earlier in development, and updates often include features asked for by the users or that are seen to be useful. Of course, it’s often not how it’s seen by you or me, but if they want business, they have to give the customers what they want.
I do wish software companies (Specifically the publishers) put more effort on a more stable and efficient product, but the current view seems to be release the software and patch it later. That’s especially prevalent in gaming, and rarely do patches fix a broken game to a fully playable state.
My wife runs Windows 7 on her Mac for one particular piece of software that is not available on the Mac. Consequently, she only uses that OS once a week or maybe twice a month. It’s not unusual for her to have a couple hours of updates to load every time she boots into Windows. By contrast her Mac sees an update about once a quarter. The experience depends on the OS. 🙂
Why? Because you want compatibility with other people, and they want more features. One can only design things right if the target stops moving. Most software is like a pedestrian bridge that was “upgraded” so that 5 ton trucks can cross it: a hodgepodge which is easy to hack into.
If one could only throw away backwards compatibility the software stack would be considerably simplified. (Like knocking down the original church to build a cathedral). To some extent that’s what Apple does. Try running your 5 year old PowerPC software on Lion… Microsoft prefers to allow you to run your 30 year old software.
Throw away all compatibility and you’d be quite happy with an AtariST. No need to ever upgrade. No useless features. A fast editor (Rédacteur) which was used by newspapers in France.
While it is annoying to be on the receiving end of continual updates, other than the comptibility issues mentioned the above poster, there sre the deliberate acts of vandalism caused by the morally destitute who insist on hacking into systems to cause as much havoc as possible, or to steal business ideas/money/what have you. The majority of my updates are security patches which wouldn’t be necessary if it wasn’t for these hackers, so I don’t get so mad at Microsoft who are only responding to peoples abuse of their perfectly functional systems.
As for them getting the security right the first time. just remember:
“What the mind of man can invent, the mind of man can circumvent.” – Triplanetary, E. E. “Doc” Smith.
Since new technology is being invented all the time, I can’t see that *anyone* is *ever* going to get it right the first time!
@Ainslie: that quote isn’t quite right. It didn’t make sense to me that it would be in “Triplanetary” – the only thing it would make sense with would be the “Golden Meteor”, the inadequacy of which (inability to resist counterfeiting, specifically) wasn’t really discussed until “First Lensman”. The latter contains the following:
“What physical science can devise and synthesize, physical science can analyze and duplicate.”
Also, “Second Stage Lensman” contains (Kandron, speaking to Alcon):
“What we could do, we have done; but what science can do, science can circumvent.”
“Children of the Lens” also contains a slightly different phrasing of the same statement as appeared in “Second Stage Lensman”.
That’s the closest I can find to your quote in the Lensman series.
Clearly Smith was acquainted with the _concept_ of measure and countermeasure. The exchange between Seaton and Tammon in “Skylark DuQuesne” makes that clear:
—- quote —-
“Oh, I didn’t mean that,” Seaton broke in. “The concept is incomplete. In
several ways. For instance, if fourth dimensional translation is used as a
weapon, you have no defense against it.”
“Of course there’s no defense against it!” Tammon defended his brain-child
like a tigress defending her young. “By the very nature of things there
can’t be any defense against it!”
At that, politeness went by the board. “You’re wrong,” Seaton said, flatly.
“By the very nature of things there has to be. All nature is built on a
system of checks and balances. Doing a job so terrifically big and so
brand new, I doubt if anybody could get the whole thing at once. Let’s go
over the theory again, together, with a microscope, to see if we can’t add
something to it somewhere?”
—- end quote —-
But if he ever used the exact words you cited, as far as I can tell, it wasn’t in either of those two series.
There are some (presumably erroneous) references to that as a quote from “Triplanetary” online.
I can find a lot of similar juxtapositions of “invent” and “circumvent”, most often with regard to security or cryptography/cryptology, but so far, I haven’t seen anything that looks like the first use of those terms in opposition.
(I’ve read the Lensman series a number of times in the decades since I first read it at maybe age ten, so while I don’t quite have it memorized, it doesn’t take a huge incongruity in an alleged reference to it to catch my attention.)
As to the issue of updates, as others have pointed out, Windows seems to follow the approach of very high backwards compatibility (at the application level, not necessarily for drivers!), which means it carries a lot of baggage, not all of which receives consistent maintenance. At the same time, for many years it emphasized ease of use (in the sense of being only a few clicks away from gratification) over security; only long after the threat became serious did that start to be reversed, by which time a lot of design choices were made the altering of which would have been visible either directly to the user, or as application incompatibility. On top of that, Windows lacked the obsession with perfectionism and consistency that Mac OS X has mostly had, the sheer technical expertise that Sun had with Solaris, or the “many eyes” of the various open source OSs like Linux and the BSDs. And yet, for reasons more to do with business than technical merit, Windows has been the volume leader (and therefore the main target of various attacks).
There are trade-offs. _All_ OSs need some updates. _All_ OSs that are around long enough eventually have to set some limits to backward compatibility, so all of them will eventually break something. Most of the time, between the end-user desktop leaders Windows and OS X, I prefer the choices that were made in OS X. The OS updates there aren’t nearly so frequent, and even .0 releases are mostly usable. OTOH, there’s perhaps more _application_ breakage on major version updates. Of maybe four hundred apps (mostly free) I’ve loaded, at least ten percent broke from 10.5 to 10.6, and maybe more than that from 10.6 to 10.7.
Part of the problem seems to be that early access requires an at least $100/year subscription, and some interpretations of the NDA for early access discourage developers working on supporting the next OS release from doing large enough scale beta testing to really be ready by the time the next OS comes out. So even though the OS itself may be (mostly) quite usable, application support is inconsistent, esp. away from the most actively developed apps. Another problem may also be what APIs an OS developer commits to as stable vs those that app developers want to be able to use to do the full range of things (some of which may include extending system behavior) they want to do. The more an OS exposes as stable, the more baggage it carries. But if an app uses unstable interfaces, it risks breakage at any update, esp. any major update.
Windows or OS X consist of _millions_ of files, and hundreds of millions of lines of code. (there are over 29 million files on my Mac right now, although a lot of those are additional apps, documents and other data, etc) No matter what choices were made, there will be pain involved to keep anything that complex reasonably secure and reliable under changing threats, evolving standards, etc. (lest those numbers sound incredible, even my _iPhone_ has over 3.2 million files, although that includes 105 additional apps and over 2600 songs; not that many years ago, few other than Clarke or Asimov would have imagined being able to carry around something with that much sheer capacity)
@ R. Hamilton
I stand corrected. I simply found a quote similar to what I intended to say on the net and didn’t check to see if the source was correct. I merely thought I was misremembering it. It was simply the meaning I was after, which seems to me to be a pretty good summation of what happens.
Perhaps it is simply an old adage rather than a direct quote. It is far too many years since I read Smith for me to remember for sure whether it was in any of his books or not, and my memory is sketchy at the best of times. It wasn’t my intent to upset anyone.
I always turn the updates off, and never have any problems at all. So it IS an option, even if Microsoft would like you to believe it is not. And of course, you can also do the updates manually, only a small portion of all those updates are actually required. (Usually the security patches and stuff.) That way it takes a bit of your conscious time, but nowhere near as much time as allowing all the updates to install would.
Good luck!
(Said the green lady from the kimono triplet at WFC Calgary.)
Thank you… and it was a lovely kimono!