From a Software User's Perspective

Thank you for reposting http://www.microsuck.com/content/whatsbad.shtml.

Mention Not :)

I believe that this is reiterated every time Microsoft release a new Windows Version or Service Pack.

If you don't like it you can always stick to your fully functional current Operating System. I know of a few people who still use Windows 98. I certainly won't be upgrading to Vista when it is released. Until such time as I can be lured to a product with the addition of relevant features there is no way that I will dare touch the product.

It wasn't until well after SP1 was released for Windows XP that I started using it, and I was fixing Windows XP machines for a living before SP1 released [grin]

Also, for what most people use in Office features, they can get from Office 97 or Open Office. That doesn't mean that companies aren't going to buy the new Office 2003 suite though, even if they will never use any of the extra features.

Anyway.. who cares..

Wow, you have uncovered their secret plan!!111 Because everyone know that when you add more drivers, fix bugs, adds the ability to customize, adds lots of effects, provide various frameworks and applications then the size of course shrinks and it will be faster, basic logic.

how are you meant to add more features and take less space..

that aside, I am posting from a p3 900 laptop (not my main machine!), it came with 2000, I upgraded to xp for free (microsoft partner, I have about 4 xp licensces left!), and it takes up minimal extra space, and runs faster.

I am sorry, but saying what you are saying is almost as good as people who say switch to linux, it is more reliable... or the apple people who say the same based on reliability.

if anyone quotes this post or wants to say anything, I will gladly say much more on the topic, I just am tired and wont if no one is interested in my oppinion!

Frankly though, everyone buys new appliances, new tv's, new games consoles, why not new computers, yes they may be expensive, however it is life!

I upgrade every 6 months or so (I build computers, and often have spares that I cant sell, so I sell what I can at minimal loss on ebay, then keep the spares!) My current pc is a 64bit 3.0ghz pentium 4 HT, 250gb hard drive... I probably can not run aero, but that wont bother me, I will run everything else of Vista.

the last windows was xp which is already about 6? years old, look at anti virus or other software, you get new versions almost every year! people want new features when you get a new OS, frankly these are some of the side effects, you need new hardware! its a tradeoff, but hardly a bad one.

and hows it rubbing intels back? windows does run on AMD's you know? (and via, plus a few others!)

Bloatware and the 80/20 Myth.

I'll bet you think I'm going to write one of those boring articles you see all over the net bemoaning "bloatware". Whine whine whine, this stuff is so bloated, oh woe is me, edlin and vi are so much better than Word and Emacs because they are svelte, etc.

Ha ha! I tricked you! I'm not going to write that article again, because it's not true.

love that link!

Quote:

Original post by Arild Fines Bloatware and the 80/20 Myth. I'll bet you think I'm going to write one of those boring articles you see all over the net bemoaning "bloatware". Whine whine whine, this stuff is so bloated, oh woe is me, edlin and vi are so much better than Word and Emacs because they are svelte, etc. Ha ha! I tricked you! I'm not going to write that article again, because it's not true.

Excellent article.

I like RegClean... =haha

Quote:

Original post by Arild Fines Bloatware and the 80/20 Myth. I'll bet you think I'm going to write one of those boring articles you see all over the net bemoaning "bloatware". Whine whine whine, this stuff is so bloated, oh woe is me, edlin and vi are so much better than Word and Emacs because they are svelte, etc. Ha ha! I tricked you! I'm not going to write that article again, because it's not true.

I don't really agree with this article. The computer I'm writing from runs Windows 98 so you might need to take this with a grain of salt.

I remember a few years ago (2001-ish) my parents wanted to upgrade IE for compatibility with some sites that weren't liking the older version. Download small setup utility which proceeds to download 64 MB. Netscape (they use both) also needed an upgrade. I don't think Netscape was quite as large, but it was in the 10's of megabytes. I ran Opera on the same computer. If I remember correctly, its download was less than 10 MB (I downloaded the new version today which weighs in at 3.65 MB) and included mouse gestures which is a feature I love that was (is?) absent from those two larger browsers.

Today, I do most of my document editing using Latex/AMStex. I do this mostly because it allows me to easily embed mathematical formulas, but also because it's my experience that pdf/dvi's are rendered correctly on a wider variety of systems (running various versions of various readers) than Word documents (which some professors require). Now, disk space wise, it's a somewhat hefty 161 MB, but that's not my complaint. I edit the documents in EditPad Lite which starts up immediately and has snappy response times. Compiling the .tex file doesn't take noticeably longer than saving a Word document. Compare this with Word which has a noticeable startup time, can behave sluggishly even on decent hardware, and is lacking features I want (easy math formulas and portability of format).

So, sure, I'll agree with the article that the 80/20 thing is a myth. However, that doesn't change the fact that there are smaller/snappier programs with more features that I actually use, features lacking in the larger/more sluggish programs.

Quote:

Original post by Way Walker I don't really agree with this article. The computer I'm writing from runs Windows 98 so you might need to take this with a grain of salt.

More like a bucket.

The point of the article is not that large application images indicate a wealth of available functionality that's there in case you ever need it. The point of the article is that, prorated against the cost of resources for both user and developer, and against the cost of delay to minimize resource consumption to both user and developer, the eradication of "bloat" is an unprofitable obsession.

You have refused to upgrade your machine/OS because you feel that what you currently have continues to serve you well. Good for you! You are not the mean case. The majority of people who comprise the audience for commercial software (free or at cost) upgrade between every 3 and 5 years, and that upgrade increases the available resources as well as the adeptness of their OS at maximizing resources (my buddy has Windows XP running on a P3 500 MHz or so, with 64 MB RAM, and it does most things quite well).

So while your anecdotes are interesting, they have virtually nothing to do with the argument of the article.

Quote:

Original post by Oluseyi Quote: Original post by Way Walker I don't really agree with this article. The computer I'm writing from runs Windows 98 so you might need to take this with a grain of salt. More like a bucket. The point of the article is not that large application images indicate a wealth of available functionality that's there in case you ever need it. The point of the article is that, prorated against the cost of resources for both user and developer, and against the cost of delay to minimize resource consumption to both user and developer, the eradication of "bloat" is an unprofitable obsession. You have refused to upgrade your machine/OS because you feel that what you currently have continues to serve you well. Good for you! You are not the mean case. The majority of people who comprise the audience for commercial software (free or at cost) upgrade between every 3 and 5 years, and that upgrade increases the available resources as well as the adeptness of their OS at maximizing resources (my buddy has Windows XP running on a P3 500 MHz or so, with 64 MB RAM, and it does most things quite well). So while your anecdotes are interesting, they have virtually nothing to do with the argument of the article.

here here! or is it hear hear!

anyway, I like what you said!

Quote:

Original post by Oluseyi The point of the article is not that large application images indicate a wealth of available functionality that's there in case you ever need it. The point of the article is that, prorated against the cost of resources for both user and developer, and against the cost of delay to minimize resource consumption to both user and developer, the eradication of "bloat" is an unprofitable obsession.

Then I think the article misses the point behind most complaints I hear about bloat. The software gets larger, remains sluggish inspite of better hardware (and is thus even more sluggish on older hardware), and doesn't offer noticeable improvements over older versions. I upgrade simply because older versions are no longer supported (I'm coming up against more and more cases where Win2k is the minimum version of Windows supported), not because they offer anything new for me. Like the article says, I'm sure someone out there is using those new features, but I don't know any of them and I don't like to pay the price for it.

One of the design principles behind C++ was "you don't pay for what you don't use". I think this hurt C++'s design a bit by adhering too closely to this idea, but, overall, I think it's a good idea and would like to see it more in "bloated" software.

if all software providers accounted for the lowest end PC there would never be any progress.

Quote:

Original post by Oluseyi (my buddy has Windows XP running on a P3 500 MHz or so, with 64 MB RAM, and it does most things quite well).

Please tell me how he's done it. The tweaks for that would help me a bunch. We have bigger machines (P4 1.6ghz 256 meg machines) that are sluggish at that amount of RAM. If he got it running well like you say, these other machines should be able to run much much better.

Quote:

Original post by flounder if all software providers accounted for the lowest end PC there would never be any progress.

They don't have to, but a word processor shipped with a modern computer that doesn't take 30 seconds longer to load and appears to offer all the same features as the word processor I had back in 95 would be very nice.

What I would like to see in a mainstream OS is installing and setting up by default to limited funtions. Want networking? Pop in your install disk and check check off the networking serivces. Want a media player? Pop in the install disk and install the media players. Have the installer designed to walk someone that only knows "I press the big button, hit this icon, and suddenly I'm on this interweb thingy" through the processes. State what each function does in as simple terms as they can, include lots of information. Let it take hours to install if needed, letting little old granny actually LEARN something about their OS. A lot of problems that users have with windows seems to stem from people being able to abuse holes in unneeded services and features that are on by defaut (or were before the newer service pack)

The article that someone posted there is nice and all, but misses a lot of points. If version 1 of the software did ok, ran fast, but lacked just a handful of features, or simply had a bad interface that is all corrected somewhere around version 6 or so, why is version 6 taking up 10 times the space, and not running any faster on better hardware? Just because memory costs a fraction of what it did before isn't a very good reason for developers to just use whatever with no worries about how much space it takes.

I agree that they shouldn't take an extra 6 months to cram everything into 2mb, rather than 20, but still, the increase of install sizes is just getting rude!

Quote:

Original post by Way Walker Then I think the article misses the point behind most complaints I hear about bloat. The software gets larger, remains sluggish inspite of better hardware (and is thus even more sluggish on older hardware), and doesn't offer noticeable improvements over older versions.

Let' examine each of those claims/complaints, shall we?

1. The software gets larger. According to what metric? Size of the installer? Size of the installation media? Size of the program's directory on disk after installation? Do we account for an increase in included assets like icons, images, templates, etc?
   
   Without metrics and rules for obtaining them, this claim is ambiguous. It pretty much comes down to "gut feeling," but nobody ever has a positive gut feeling about technology, really, so how objective is this assertion/complaint?


2. The software remains sluggish in spite of better hardware. How is this sluggishness measured? Are we relying on stopwatch-timed measurements for various activities? Do we factor in whether a particular operation (like loading) does more or less in either the newer or older version? And how much better is the hardware, compared to the increase in the load that the software represents? If the hardware is three times better but the software is five times more complex/feature-rich/"bloated", there should be a net decrease in perceived performance.
   
   Again, it seems like this complaint is based on psycho-suggestive perception rather than emperical data. Which is a classic hallmark of FUD.


3. The software doesn't offer noticeable improvements over previous versions. Who's doing the noticing? Microsoft Word 2003 has over 2000 functions and features. Word 6.0 had, I believe, about 150. Yet many people use fewer capabilities today than Word 6.0 had in 1995 or whenever. Does that mean the software didn't improve, since they didn't notice/need the new functionality?

People complain. It's what they do. And while there is valuable feedback in their complaints, it's not in taking them verbatim - because people often don't know nearly enough to make expert assessments. The key here is managing user perception: how can developers make user feel that the software is faster and/or more responsive? In recent years we've seen the introduction of various measures that slow the user down from starting work immediately, but present data or workflow management features almost instantaneously so users can feel productive; splash screens that list the ridiculous amount of things they are supposedly doing, so users counter there "Why is this taking so long?" with "Shit, I never realized it did so much work in the background!"; reduced latency in UI prompts and indicators; and much more.

People complain, and often their complaints are driven/fueled by misinformation - deliberate or no - stemming from science fiction fantasies and the barely literate ruminations of "technology writers."

Quote:

I upgrade simply because older versions are no longer supported (I'm coming up against more and more cases where Win2k is the minimum version of Windows supported), not because they offer anything new for me. Like the article says, I'm sure someone out there is using those new features, but I don't know any of them and I don't like to pay the price for it.

Then stick to what you're using. Don't upgrade. Buck the trend. But accept the consequences of doing that: the death of support. A company can not support all of its products in perpetuity. You wouldn't take a Model T to a Ford service center and expect them to tune it up, no questions asked. You'd take it to a classic car specialist, or someone else who can locate or custom fabricate the parts and components your century-old vehicle will require.

Well, shocker!, software ages, too. The people who wrote and can thus best maintain Windows 98 - which is like a Model T compared to XP's Explorer - have moved on, many out of Microsoft. Apple doesn't provide service options for the Lisa any longer. Hell, it doesn't provide service options for the Tangerine iMac G3! HP doesn't stock/sell toner for the LaserJet II any longer. In each of these cases, you'd turn to a specialist without a complaint.

Want to keep running DOS 6.22? Become/find a specialist.

Quote:

One of the design principles behind C++ was "you don't pay for what you don't use". I think this hurt C++'s design a bit by adhering too closely to this idea, but, overall, I think it's a good idea and would like to see it more in "bloated" software.

Considering how spectacularly poorly suited C++ is to modern application development, even after the 1997 retrofitting of pseudo-modern components onto its aging core, and the on-going efforts of projects like Boost, I'd much rather general purpose technologies didn't try to emulate anything in C++, thank you very much.

Quote:

Original post by Oluseyi HP doesn't stock/sell toner for the LaserJet II any longer.

oh crap, dont tell me I have to chuck mine away now! (lol, I seriously have one of these in my cupboard I think, I got it from a office that chucked it out, Printed about 10 pages, worked fast and great, then it said "please service" or something like that, no idea why! and now it has been in my cupboard for 5-10 years! its gigantic and maybe I should get rid of it!

Quote:

Original post by Oluseyi Let' examine each of those claims/complaints, shall we?

I'm game.

Quote:

The software gets larger. According to what metric? Size of the installer? Size of the installation media? Size of the program's directory on disk after installation? Do we account for an increase in included assets like icons, images, templates, etc? Without metrics and rules for obtaining them, this claim is ambiguous. It pretty much comes down to "gut feeling," but nobody ever has a positive gut feeling about technology, really, so how objective is this assertion/complaint?

I'll take the article's analysis to be fairly objective on this. Take Excel as a typical program and note that its absolute size has increased while size relative to hard drive capacity and cost of storage have gone down. I still feel justified in asking why the absolute size is going up even though I nor those I know have found any of the additions useful and sometimes have to go searching for an old favorite that got moved (or in rare cases removed).

Quote:

The software remains sluggish in spite of better hardware. How is this sluggishness measured? Are we relying on stopwatch-timed measurements for various activities? Do we factor in whether a particular operation (like loading) does more or less in either the newer or older version? And how much better is the hardware, compared to the increase in the load that the software represents? If the hardware is three times better but the software is five times more complex/feature-rich/"bloated", there should be a net decrease in perceived performance. Again, it seems like this complaint is based on psycho-suggestive perception rather than emperical data. Which is a classic hallmark of FUD.

Again, I haven't actually taken a stopwatch to a computer, but the various versions of Word has always felt about equally sluggish on every computer I've used it on since the late 90's (well, more sluggish when Word is of a newer vintage than the hardware). Does stopwatch data exist? Why would it feel equally sluggish if it isn't? (I do regularly hop between different hardware and different versions, so it's not "back in my day...") Even if it is "just" my perceptions and it's actually getting faster, then it's still not "perceptibly" faster.

As for factoring in whether it's doing more or less work than before, I'm saying that's only relevant if I see some benefit from that "more" (i.e. I use that new feature) or that "less" (e.g. it's faster without that feature). I'll pay a little in loading time for something I'll use and I don't care about features I don't pay for; I don't like paying for features I don't use. Also, I'm willing to pay more in loading time than I am in responsiveness while I'm working.

Quote:

The software doesn't offer noticeable improvements over previous versions. Who's doing the noticing? Microsoft Word 2003 has over 2000 functions and features. Word 6.0 had, I believe, about 150. Yet many people use fewer capabilities today than Word 6.0 had in 1995 or whenever. Does that mean the software didn't improve, since they didn't notice/need the new functionality?

Who? Me. Friends. Fellow students.

I'd say the software didn't improve for them if they don't find a use for the new functionality. I'd say the software got worse for them if it takes more space (measuring relative to hard drive capacity) or requires better hardware to get the same performance and they don't find a use for the new functionality.

Quote:

People complain. It's what they do. And while there is valuable feedback in their complaints, it's not in taking them verbatim - because people often don't know nearly enough to make expert assessments. The key here is managing user perception: how can developers make user feel that the software is faster and/or more responsive? In recent years we've seen the introduction of various measures that slow the user down from starting work immediately, but present data or workflow management features almost instantaneously so users can feel productive; splash screens that list the ridiculous amount of things they are supposedly doing, so users counter there "Why is this taking so long?" with "Shit, I never realized it did so much work in the background!"; reduced latency in UI prompts and indicators; and much more.

Well, one way to make the user feel that the software is faster and/or more responsive is to actually make it faster and/or more responsive. On my parents' computer, Opera is more responsive than IE or Netscape.

All I get from the splash screen listing what it's doing is the knowledge that it's not taking so long simply because it crashed. A progress bar would be a bit more helpful, but I'll still curse it for taking so long. Offering me some functionality while loading others and reducing UI and indicator latencies is actually appreciated.

Quote:

Quote: I upgrade simply because older versions are no longer supported (I'm coming up against more and more cases where Win2k is the minimum version of Windows supported), not because they offer anything new for me. Like the article says, I'm sure someone out there is using those new features, but I don't know any of them and I don't like to pay the price for it. Then stick to what you're using. Don't upgrade. Buck the trend. But accept the consequences of doing that: the death of support. A company can not support all of its products in perpetuity. You wouldn't take a Model T to a Ford service center and expect them to tune it up, no questions asked. You'd take it to a classic car specialist, or someone else who can locate or custom fabricate the parts and components your century-old vehicle will require.

Well, it's not so much that MS doesn't support Windows 98 as that software vendors don't. Also, I'm not complaining about the lack of support, I'm just saying that's the only reason I see to upgrade. I have to upgrade because I have to be able to open documents saved using the latest version of X. It'd be nice if the new version offered me something other than slower performance, larger absolute size on hard drive (which means larger relative size if I don't upgrade my hard drive as well), and files that no longer work with older versions.

Quote:

Well, shocker!, software ages, too. The people who wrote and can thus best maintain Windows 98 - which is like a Model T compared to XP's Explorer - have moved on, many out of Microsoft. Apple doesn't provide service options for the Lisa any longer. Hell, it doesn't provide service options for the Tangerine iMac G3! HP doesn't stock/sell toner for the LaserJet II any longer. In each of these cases, you'd turn to a specialist without a complaint.

Ya don't say? The difference is a Ford Focus blows a Model T out of the water for daily use. Using a lab computer running XP isn't significantly different from using my home computer running 98. And, seriously, their excuse has to be better than, "It was easier for the new guy to write his own version than to learn the old guy's code".

As for cars aging, I drive a 1990 Oldsmobile that I have no trouble getting serviced. I know people with older vehicles that don't need a specialist. Heck, Oldsmobile doesn't even exist anymore, so I don't buy the "The old guy quit, all we got is the new guy" excuse.

Quote:

Considering how spectacularly poorly suited C++ is to modern application development, even after the 1997 retrofitting of pseudo-modern components onto its aging core, and the on-going efforts of projects like Boost, I'd much rather general purpose technologies didn't try to emulate anything in C++, thank you very much.

Hey, I'm the first to criticize C++, but to say it had no good ideas is short-sited. In fact, a lot of software does the "you don't pay for what you don't use" to a limited extent by asking you which portions you want to install and offering plugins for some features. I'd like to see even more of this.