Wednesday, February 11, 2009

Koala Flags Down Cyclists to Quench Thirst in Australian Heat Wave

I don’t know whom to credit for these images, sent to me by a friend in South Australia. I will gladly attribute them properly, it the photographer contacts me.

It has been extremely hot in south-eastern Australia during the past few weeks, especially in Victoria (where the tragic, deadly bushfires raged last Saturday, 7th February) and in South Australia.

Below are some photos showing what koalas are having to do to get a drink. The photos with cyclists were taken on Tuesday night, 3rd February 2009, on the old “Eagle on the Hill" Highway (near Adelaide, South Australia).


 
UPDATE:
Here’s another koala, a female called Sam, rescued by a fireman the following weekend in Victoria. Sam was so thirsty that she drank three bottles of water:

Tuesday, February 10, 2009

Bushfire catastrophe in Victoria, 07 February 2009

Last Saturday afternoon I reported that down in the south-eastern corner of Australia, in the state of Victoria, we were the hottest day ever on record, and that the expected bushfires were raging all over the state.

The Bunyip State Park blaze has been fanned by gusty winds. (LINK - AAP: Simon Mossman, file photo)

It saddens me to report that as I was writing that blog post, another tragic record was being broken: the most lives ever lost in Australia due to a bushfire inferno (several hundred souls lost, and the count is still increasing as the ashes are combed). It is now classified as Australia’s worst ever peacetime natural disaster.

Many people live in towns and villages an hour or so by road from downtown Melbourne, enjoying the beautiful Australian bush while being quite close to a major metropolis.

But mother nature had this terrible event in store, with hot gale-force northwesterly winds funneled in to the Victorian region after traveling across the hot continent, and on a day of record temperature. Whether caused by lightning strikes, accident or arson, the fires roared through the tinder dry forests at speeds that caught out even the best-prepared citizens.

One eye witness described a fire front first appearing some twenty kilometres away, then reaching his location about two minutes later. TWO MINUTES! Others said that the front approached like a tsunami of flame ten or twenty metres high.

Embers were reported settling down up to ten or fifteen kilometres ahead of the fire fronts. People found themselves suddenly and totally unexpectedly surround by raging fires. A few were lucky to survive where they were, but too many poor souls weren’t.

Some tried to flee by car, but they had no hope of outpacing the fires and were incinerated as they drove.Victoria's killer fires, 07 February 2009, No. 105

There were shocking images of the shells of their cars, showing the alloy wheels having melted and spread across the road. One driver who was lucky to survive said that his timber truck had melted, not burned!

The Australian Broadcasting Corporation (ABC) has coverage of the tragic bushfires here.

Victoria had two earlier bushfire disasters in the summers of 1983 (“Ash Wednesday”) and 1939 (“Black Friday”) but the firestorms of 7th February this year were even worse than those.

Please spare a thought and say a prayer for all of the victims and their families.

You couldn’t stop this inferno with 10,000 fire trucks and 1,000 helicopters:

Is it POSSible (or PROSSible) that the great FOSS debate is missing an important point?

David M. Williams (one of my fellow writers at iTWire) has just posted an opinion piece that got my grey matter going, see Free software isn't freeware: why Linux and FOSS have a higher standard

You’ll find some interesting, on the whole supportive, comments to David’s article here.

I thought that thee was an additional aspect of open source missing from his arguments, and missing also from just about every other thing I’ve read on the FOSS topic, including those by the arch-advocate Richard Stallman, so I added these comments on iTWire

You haven't quite covered all the bases regarding "free" and "gratis" and "non-gratis" and "open source" in my opinion.

Let's have a pseudo-hypothetical example. Apart from other gratis apps that I generously offer, suppose I have a non-gratis product "Fantastic App" -- gotta make a living somehow, so why not charge for at least one of my apps?

I've slaved away developing over many months, and keep coming out with minor enhancements (let's call these "releases') and less frequently major enhancements (let's call these "versions"). If you are willing to agree with the license and usage terms for FantasticApp and pay a license fee for any given version of it, then I let you have it together with all the source code to use however you like throughout your organization. All bug fixes and minor releases incur no extra fee, but there's an upgrade fee if you choose to upgrade to a newer major version.

Since you have all the source then surely FantasticApp is "open source" is it not? If I get run over by the proverbial Bourke Street bus (you probably don't know that we had double-decker buses in Bourke Street, Melbourne, some decades ago), there's no issue since you have all of the code.

But it's proprietary because the license states that you cannot share it with anybody outside your organization. Nobody forced you to purchase the license in the first place. You were free to try to get the same or a similar app elsewhere, presumably agreeing to license my app because it was the only one of its sort or the other were in some way unsuitable for your needs.

I would intend to demand payment from and/or sue any organization that obtained the source for FantasticApp without paying the appropriate license fee, or released to other organizations who hadn't paid up.

So this is a piece of non-gratis "proprietary open source software" (or "POSS") is it not? Or maybe it would be more accurately labeled "proprietary restricted open source software" (or "PROSS"). What's wrong with all that? Surely we aren't expected to develop apps like FantasticApp for no financial return if we want to make a living from them from at least some of our efforts, that seems absurd to me.

Richard Stallman, eat your heart out over this one!

And what are your opinions on my POSS/PROSS concept?


Saturday, February 07, 2009

Hottest day ever in Melbourne!

It’s Saturday 7th February 2009, and the temperature reached 46.4 Celsius, the hottest day on record for Melbourne (and the entire state of Victoria).

Gale force northwesterly winds became scorchingly hot as they baked while moving across the oven that is the heartland of Australia. … Bushfires all over the state, loss of property, all as expected.

Here’s the thermometer I carried back from one of my work trips to the IBM Development Lab in Rochester Minnesota, where it now sits on the fence in our back yard.

For Yankee people not used to thinking in Celsius mode, here’s a mid-afternoon shot:
Hottest_day_Melbourne_07_February_2009_#3  

That’s around 113 degrees Fahrenheit. A tad on the warm side, eh?

I’ve experienced a winter in Minnesota, so I know what it’s like when the needle points toward the 9 o’clock position, but here in suburban Melbourne the needle will barely get over to the left side of the scale.

UPDATE:
I forgot to mention yesterday that while southeastern Australia (South Australia, Victoria and New South Wales) has been in severe drought and sweltering in oven-like conditions, the northern sub-tropical parts of the country have been awash under meters of floodwater in this year’s monsoon.cyclone season.


A squeaky wheel runs amok in ultra-hot hot weather (all about Lotus Notes 9.0)

Thanks to the great meeting place that PLANET LOTUS has now become, I came across the following post by Glenn Irvine: What I’d Love to See in Notes 9.0 (go take a look at it)

Let it be clear from the outset I have nothing whatsoever against the feature changes that Glenn is requesting. It’s just that when he talked about “Notes 9.0” I expected it it the be about, well, Notes 9 as a whole – but it only happened to be about several usability shortcomings in the mail application in Notes.

So I made the following comment (maybe because it’s around 10 AM and already warming up. The Bureau of Meteorology predicts a real scorcher here in Victoria today. They reckon it will reach 44 degrees Celsius or more, which would make it the hottest February day on record down here, and very ominous for bushfires. (UPDATE: see what actually happened here.)

We had a couple of similarly hot days near the end of January. People, animals and plants are under considerable weather stress in our corner of the continent. The last rain of any significance was in December last. The weather was quite cool then, but it has warmed up with a vengeance recently.

A day or two ago, ex-Collingwood footballer Peter Daicos, when putting out his wheelie bin in North Balwyn (just a few kilometres away in the heart of suburban Melbourne) was bitten on the toe by a venomous red-bellied black snake. (Talking about snake, scientists have found in Colombia a fossil 13-metre giant snake, as long as a school bus, which used to devour giant turtles and primitive crocodiles. Scary.)

Another sign of the unusually warm stretch we’ve had here in south-eastern Australia is the fact that many of the less hardy deciduous trees have been dropping their leaves, months before the stat of our southern autumn.

Maybe this has put me a bit on edge, but I just commented on Glenn’s story as follows. Let me know if you agree, or otherwise:

- - - - - - - - - - - - -
What disturbs me a bit about the content (but sometimes only the title) of posts like this is that all too often they’re restricted to talking about desired enhancements in apps bundled up with Notes. Often the comments are restricted to Notes Mail (as in your case above, Glenn) — or to iNotes or Sametime or one of the other IBM apps.

Understandably, IBM people (like our famous and most gentlemanly leader Ed Brill) often talk about specific enhancements in their own apps. Hey, I worked for IBM for a long time, and know that it’s hard enough to keep up with what IBM is announcing much less what happens outside IBM.

Many of the vast community of IBM business partners and independent ISVs, not to mention all those thousands of Notes customers big and small alike who roll their own apps (and some of whom have a vast inventory of such apps), do like what they see being added to or fixed in in Notes Mail.

But in many cases they also want major new generic enhancements in Notes. A perfect example of this would be the grouse new Xpages capability that’s just been released in Notes 8.5 for us all to savor.

Notes remains a fantastic application platform [for certain classes of apps] as we all know. While there have been many outstanding mainly server-related enhancements in the past few years (major compression improvements, DAOS, etc) it’s not quite as common for a startling application-building enhancement like Xpages to appear.

So what would I myself talk about when I say I’d like to see something new in Notes 9 or Notes 10?

How about a few REALLY big hitters, for generic app design and development? For one, the ability to design an application once and have it behave exactly the same in the Notes client and a web browser? (Not asking for much, am I?)

IBM in an earlier release used to make available what I’ll term “some trusty old Lotus Notes applications” which you’ll find listed here on my website: http://asiapac.com.au/Links/NotesDomino.htm#TRUSTY_OLD_LOTUS_NOTES_APPLICATIONS

They’re quite old and rusty pre-Domino apps (so are not web enabled), but my point here is that one thing they did at that point in time was to let people see that Notes is note just about mail (as important as mail still is), but that Notes is really very good for building a vast range of other application types.

The old hands at Notes (customers and others) were very worried that IBM had lost its way in the early part of this decade, but needn’t worry now since it’s apparent that IBM is solidly behind Lotus Software including Notes.

A recent extremely promising move from IBM is the decision to get behind OpenNTF.org and help push out the message that Notes is a great application platform, and that there are lots of free apps available from OpenNTF (and from a range other sources too, for that matter)

In summary, what I would have called this post is “What I’d Love to See in Notes 9.0 Mail” because there are LOTS of other bigger things I’d like to see in Notes 9 (and 10 and 11 and ad infinitum) as a whole!
- - - - - - - - - - - - -


Friday, February 06, 2009

An oddity in Windows Live Writer

As mentioned in the previous post (about OpenNTF.Org), I now use the Writer available under Windows Live to create and update posts to my blogs.

I decided to check out the Writer web site for updates, and installed the latest version (build 14.0.8050.1202) which has even more nice features. The upgrade all went fine.

That done, I started researching a problem that has cropped up a few days ago, with Windows inscrutably refusing to go into Standby mode.

It’s very irritating. Now, instead of quickly powering down into standby the system nearly always hangs displaying the “Windows is now shutting down” panel. My only option then is to do a hardware reset in order to get going again.

Some software that I installed or patched in the last week or so is probably the culprit. After a little research, I  decided to uninstall certain products that I suspected might be the cause of the hangs. (Windows power configuration issues like this can be frustratingly difficult to sort out.)

So I opened the Control Panel, and got the following result (only the top few lines are shown). Notice the oddity, near the top?

 Windows_live_writer_oddity 
  (Click to show a larger image)

I’ve never seen the likes of it before, in any version starting with Windows 95.

An examination of the Windows registry (by searching for “LiveWriter”) detected the minor but amusing gaffe made by somebody at Microsoft:

Windows_live_writer_oddity_registry_ProductName_entry
(A quite unusual registry key value)

HTML rules, it has now infested even the registry! … And I thought it was only me who did silly things like this.


Messing around with OpenNTF.Org’s sign on and download process (updated)

Don’t get me wrong, I reckon that OpenNTF.org is an absolutely fantastic boon to the Lotus Notes/Domino community. There are lots of excellent tools, frameworks, utilities and applications available for download at the best possible price [free, as in beer].

But right nowOpenNTF.org_reset_password1 I cannot download anything. Frustration… It accepted my long-time password without complaint, but still shows my status as “not logged in” and prevents me from downloading anything.

So I request a new password, which it tells me will arrive shortly, but I have to wait maybe fifteen minutes for it (didn’t take an exact measurement of the elapsed time). Having to wait so long was another frustration, such things should happen within seconds, a minute or two at the very outside.

I signed in using the new password, but it still showed me as being “not logged in” (highlighted in yellow in Figure 1) so I couldn’t download anything. Therefore I repeat the sign-on process a number of times more, sometimes deliberately entering the wrong password and having it knocked back. … Swearwords flowed liberally!

After a few more minutes of puzzlement and messing around, I decide to try the same with Firefox browser, and everything worked fine this time. Then tried to work out what configuration setting (or cookie or whatever) was causing my IE-based Avant Browser session to spit the dummy and prevent downloading [even while accepting the new password].OpenNTF.org_reset_password2

I decided to give up on that quest and, not liking the cryptic new password, wanted to reset it to its old easily-remembered value.

I hoped to find a conventional “Manage your profile” or “Reset Password” function just underneath the “Register” and “Forgot Password” links in the right-hand column (where the green highlighted area is in Figure 1) but it wasn’t there.

Maybe I was just just having a bad morning or am going blind, but having signed in (under Avant Browser) couldn’t see a “reset password” facility anywhere, so it appears that I’m stuck with the cryptic password.

[UPDATE STARTS HERE] But then I look at the Firefox window again and, as shown in Figure 2, see that indeed there is an “Update Profile” option (thought there would be all along, actually).

Now I’m an old-timer in the IT industry – still think that punched cards and print-outs are the most reliable form of input/output devices devised so far!  One of the reasons I have misgivings about new fads like AJAX and cloud computing is that they’re so reliant on HTML and browsers in all their variations, and problems like I experienced this morning with cookies (or whatever it was) are really awful to experience and nightmarish for the web application designer/developer to handle elegantly.

I write regular articles for iTWire which uses Joomla and so I have to use a browser-based editor which I find quite awkward to use. And the built-in browser-based editor for Blogger.com that I originally used for writing posts in this very blog was also quite clumsy and restrictive. even compared with the nice Writely editor they acquired for  Google Docs (which they should adapt for use in Blogger.com, I’d recommend). But now I use Microsoft’s excellent free Windows Live Writer and find it far, far nicer to use for blog content management. It’s a cinch to create everything offline in a familiar GUI (very similar to using Microsoft Word, or Expression Web / FrontPage) and when you’ve finished composing your article you just click on the Publish button and everything (text, images, more) is updated on you blog with no hassles whatsoever. It’s a very nice piece of work, congrats Microsoft.

Browsers do some things nicely, but I find them not at all good – sometimes quite horrible -- for really serious transactional activities of various types. Conceived as a means of displaying web contents, they are now being force-fitted into all sorts of environments where they perform less than ideally.

Footnote:
I find the OpenNTF site (kindly hosted by Prominic) to be a little on the slow side. My impression is that the average response time is around ten seconds or so, just on the borderline of acceptability. It would be nice if it were snappier (and a better advertisement for Prominic’s Lotus Domino hosting offering -- hint, hint).


UPDATE #2:
I implied above that I might possibly be going blind. Being a thorough type, I did further research on the causes of blindness, and just by chance came across the following little gem concerning a form of blindness (in dating)…

UPDATE #3:
Strangely enough, next day the sign on problems that I had experienced under Avant Browser (a front end to IE) went away. I have no idea why.

Thursday, February 05, 2009

Coexistence of Notes 7 and Notes 8 on the same system

Over the last couple of years I’ve posted some blog articles explaining how you can install several different releases of the Lotus Notes client on a single PC.

I know from web page hit tracking that these articles are opened many times per week, and from every continent. I can picture out there the wide range of Domino developers and administrators (at IBM customers and business partners, plus ISVs), all developing and demonstrating and supporting users across the spread of Lotus Notes releases, with the common aim of having several releases of Lotus Notes client and Domino Designer coexistent on a single system.

The last, most detailed post was in April 2008, see: Steps for installing multiple Lotus Notes Client releases on a single system in which I explained exactly how you might go about installing Notes 5, 6 and 7 releases (plus the benefits and shortcomings, which I won’t repeat here).

I’ve now removed Notes R4.6 and R5 and R6 from my system, and recently have been working with only R7.0.3 and 8.0 on my development system.

All along I’ve also been testing, in a Virtual PC environment, the beta versions of Notes 8.5 and as soon as R8.5 went gold I replaced R.0 so that I can now report (using exactly the same approach as laid out in the earlier posts).

I have verified that NotesTracker operates without any drama under Notes 8.5 (both the standard and the basic client).

I was just about to launch into using DDE (Eclipse-based Domino Designer 8.5) for doing work on the next release of NotesTracker (version 5.2).

But then I came across Tommy Varland’s two recent bug reports for DDE 8.5:

  1. My first bug found in N/D 8.5 … “When you're working with an application in the new Domino Designer, and open/close the application in the Client, the QueryClose event doesn't get fired. I had to close DD to get the event to fire. This probably isn't the most serious bug, but it is annoying when working with/testing QueryClose.”
  2. Serious bug in N/D 8.5 standard configuration - NotesUIWorkspace.CurrentDocument returns Nothing … “When debugging code that fetches the active NotesUIDocument from NotesUIWorkspace, NotesUIWorkspace.CurrentDocument randomly (?) returns Nothing.”

Well, it so happens that NotesTracker relies very extensively on events like QueryOpen, PostOpen and QueryClose. So for me the first of these DDE bugs would cause me constant, excruciating pain!

Therefore for the moment I’m going to keep on using the trusty old Domino Designer 7.0.3 until DDE 8.5 is fixed up, however long that takes. I wonder what sense of urgency the IBM team has about correcting these (and any similar) designer bugs. When can we expect fixes for them?

I’ll therefore only be using the new DDE 8.5 for research into Xpages and all the other interesting new development options in release 8.5.

UPDATE: I more recently found “QueryClose Annoyances” (which IBM's Andre Guirard posted back in early November 2008) giving some insights into the less-than-perfect implementation of QueryClose in Notes 8.x Standard clients (problem report SPR #AUDITS). It seems that the completely redesigned Eclipse-based front end and the C++ back end code are not yet dancing together in complete unison, but this is understandable in a radically changed code base -- even if not quite forgivable (they should have caught any severe basic errors like this before release).

By the way, I had been wondering about using the 8.5 designer and saving the NotesTracker databases and templates with .NS7 and .NT7 file extensions, respectively. Despite various attempts over the years, I’ve never found an official Lotus Software technical description of exactly how the .NSx and .NTx file extensions work.

Can anybody point me to any such technical description? Does is even still work? That is, suppose I create (under the 8.5 designer) an application with a file extension of .NS7 or .NS5 for argument’s sake. Will I be able to open such a database under a client at R7 level or R5 level, respectively? (Let’s assume, naturally, that I don’t use any LotusScript or Formula Language feature that is unavailable in R7 or R5, respectively.)

Putting it in generic terms, if you use Domino Designer at release level “n” how many back levels will this trick work for? “n –1” or “n – 2” or what? Is this formally and explicitly described anywhere?



First impressions of Skype 4.0 – mainly very good, but a few stumbles!


I’ve been a longtime user of Skype, and over at iTWire I’ve put down my first ideas about Skype 4.0 (mainly about several details of the new GUI).

See Skype 4.0 - first impressions of some GUI and function changes


Technorati Tags: ,,

Monday, January 26, 2009

Windows 7 won’t plug this dyke, I’m quite sure

I’ve been using computers and studying their innards, software-wise, for forty years or more. The first twenty years was nearly all with various IBM mainframe and midrange commercial systems, process control systems, communications controllers and the like.

Then in the early 1980s along came minicomputers that the common man could afford to buy for personal use. I used the Apple IIe and the Commodore Amiga and later the IBM PC, all essentially with primitive disk operating operating systems (initially diskette based, later hard disk based).

The early PCs mostly had fairly crude GUIs which, as far as the the average user was concerned, only stated to get adopted on any scale when Microsoft came out with -- pirated, some would say – their own windows-style OS towards the end of the eighties. They suffered from problems such as limited memory access (the dreaded 640K barrier, and so on).

I didn’t ever use Windows 1 and Windows 2, first got my hands on version 3.0/3.10, not too long after that, version 3.11 (or “Windows for Workgroups”).

For its time, Windows 3.11 wasn’t too bad, but it suffered from a number of drastic architectural limits, like problems with 16-bit addressing of ever-increasing memory and hard disks (The problems were with handling megabytes and gigabytes in Windows 3 and Windows 9x days. Now in the early 21st century it has factored up to handling gigabyte and terabyte dimensions. Will it ever end? Not likely.)

Has Microsoft lost its way with Windows? Or have they “lost the plot” as some have put it?

There’s a lot of talk at the moment about Windows 7. Will it save Microsoft, or is it in effect just “Windows Vista Service Pack 3?” I’ve downloaded Win7 Beta 1 version and will give it a fling, in a virtual machine, some time soon, when I get a round tuit. I’ve read a fair bit about Windows 7 and have seen it demonstrated at a couple of Microsoft events here in Australia, so I”m pretty familiar with what it promises.

Where do I fit in the spectrum of opinion folders? Well, right now, having refused to install Windows Vista because I just dislike its look and feel and the irritating UAC nag prompt, I reckon that there are a few nice enhancements in Win7.

So it’s very likely that I’ll install Win7 soon after it ships, as long as Microsoft make it possible to upgrade direct from Win XP Pro to Win7. (With all the tools and add-ons that I’ve installed, Win XP still works fine as far as I’m concerned.)image

What discourages me is that, after some two decades, Microsoft appear to be making no significant move towards introducing an “OS for the future” that eliminates some long-time basic architectural shortcomings and flaws.

This has been discussed by others and I won’t repeat them here but instead let me here describe one such shortcoming, a very basic one in fact, that I haven’t seen raised elsewhere.

Take a look at Figure 1 (click to view a larger image).

I tried to mount a new disk drive. (It happens to be a TrueCrypt encrypted drive, but that’s not pertinent. It could equally well have been a USB drive or similar.)

I selected a file and clicked the “Mount” button. Next I was prompted to select a free drive letter from the list. What free drive letter? The list was empty.

The drive list was empty! And that brings out the basic shortcoming, which arises from the architectural underpinnings going way back to the original MS-DOS.

In the early days, many or indeed most PCs only had only a single diskette drive (it was something of a luxury to even have a second one). A year or two later, hard disk drives started appearing on top-end PCs, all of 10 MB or 20 MB in size, for those who could afford them (they were prohibitively expensive).

What’s the cause for the empty list of drives above? Well, these days it’s common to have quite a few drives on a PC – CD/DVD drives, internal hard disks, external USB or Firewire or eSATA drives, network drives, virtual drives, and so on. image

And the figure on the right shows how my system was just before I attempted to mount the additional (TrueCrypt) drive.

There are 2 terabytes of internal hard disk drives (some physical, some logical) fusing up drive letters C: to N:

Also plugged in was an external hard drive made up of five logical drives, two internal DVD drives plus an external one, a digital card reader with four slots for transferring my photos, and several USB thumb drives.

In Figure 2, the external hard drive letters T: and U: are highlighted in yellow, and the arrow immediately under them points to the spot where three more logical drives on this device should have appeared in the list -- but didn’t, because there were no more free drive letters, although there was no warning at all that this was the case! So naturally enough, when I attempted to mount the additional (TrueCrypt) drive it correctly showed en empty drive list.

Admittedly, for the purposes of the exercise I added more drives than I normally would ever use. I usually have four or five spare drive letters available. When I disconnected several of the devices, the list of available drives went back to a more normal situation, as illustrated in Figure 3.

image

But I’m sure you get my point: still having a single alphabetical drive letter for mounted devices is a very nasty Windows limitation.

We don’t need such a basic flaw in the 21st century, and Microsoft really should already have released a simple, elegant solution for this and other such basic flaws (emphasis on simple).

Sure, Windows NTFS has symbolic links and Vista has extended this to symbolic links junction points, but these only partially resolve the issue and are strictly for IT specialists or power users only, definitely not the average user. (Symlinks originated in the early 1980s for BSD 4.2 UNIX, and are currently found in Linux and Mac OS too.)

Let’s suppose that Microsoft decided to go ahead with some sort of fix for this issue, what might it be?

In my opinion, a neat solution would be to allow two-letter or why not multi-alphameric (letters plus digits) drive names, not just single letters of the alphabet. And without  necessarily changing anything else in file system support, in order to minimize other impacts.

For example, your marketing applications and files might be on drives MKTG1: and MKTG2: (or MARKETING1: and MARKETING2: to avoid cryptic naming and so make it even plainer to nontechnical users).

Another related bugbear of mine is that in Windows you have to carefully manage where folders and files are placed on the various available drives. IBM’s System/38 introduced an elegant solution for this storage problem way back in 1978, and the concept still thrives in its descendant the “IBM i” system (that’s the latest in a series of brand names for this family from IBM’s development laboratory in Rochester, Minnesota).

I was intimately involved with S/38 and its wonderful follow-on the AS/400, or “Application System/400”, emphasizing that it’s prime directive was to be as a superb platform for serving applications. (For the record, I happen to consider the S/38 architecture and the fundamental Lotus Notes architecture two of the outstanding widespread commercially-released IT conceptions of the 20th century.)

Among the wonderful architectural features of the S/38 and its descendants is single-level storage (see here or here or here or here for some insights).

Around the turn of the century, Microsoft was rumored to be working on some great changes to file storage, labeled WinFS. These were shelved in mid-2006, and perhaps the closest we’ve come to seeing any real change so far is Windows 7 Libraries (see Ars Technica’s overview).

The System/38 and its descendants also have libraries, but these are completely different: they are named groupings of related objects (programs, database files, queues, and any other sort of system object) that make the design, development, securing and running of jobs extremely easy to be controlled. They could be thought of sort of as “classpaths on steroids,” to make a crude analogy. But this is a different topic a very important topic, more about it some other time perhaps, since Windows PATHs and Java CLASSPATHs make up another weak point.

What do you think? When, if ever, will Windows fix basic shortcomings/flaws like this?

 

UPDATED on 27 January 2009, to rebalance some of the wording and to correct bad spots in spelling/grammar.