Lean Thinking & The Toyota Product / Product Development system goes beyond software development; it's a major shift in the way people think about product development and operations. Agile methods are quite related to Lean, and this was one of the first books (along wtih David Anderson's "Agile Management" to articulate the relationship).
I just hope more rigour and discipline can be instilled to get past the bullshit surface-level debates (how much documentation, XP = hacking?, pair programming is/isn't stupid) that really distract from the whole point of the exercise.
I would highly suggest the following to get a better understanding of how this might fit into a software context:
Michael Kennedy's book "Product Development for the Lean Enterprise"
Womack & Jones' "Lean Thinking" or "The Machine that Changed the World"
Jeff Liker's "The Toyota Way"
Toyota doesn't have a meticulously defined development methodology. Nor does it use PERT or Gantt charts to track its car development projects. Nor does it have a "project manager" to sequence & assign tasks (it relies on a "Chief Engineer" to coordinate the whole effort). Car development certainly involves a lot of written documentation, but there is no massive "requirements" document that is written up front -- there are lots of alternative requirements and design documents for sub-systems and parts and various tradeoff curves. The scope and requirements vary through the cycle of the project. There is also iteration on design, and lots of reuse of prior designs and artifacts.
Toyota designs cars in a way that many people in the Agile community would love to design software -- in an interative and incremental fashion, without the massive soul-deadening top-down process enforcement. And they deliver the best quality, time to market, and profits in the auto industry.
Lean product development is about iteration, and it's how Toyota & Lexus designs its cars, noted as the most reliable in the world. Why can't these techniques also apply to software, which is arguably even easier to change iteratively than car part designs?
And Toyota is slowly driving their competitors into the ground. I'd take another look. There's a lot more to Agile than the surface-level debates, especially if you look at it in the context of Lean product development techniques.
Six sigma is more about production processes, but Lean product development certainly has a lot in common with agile methods. It takes a bit more rigourous approach to the process though -- development certainly isn't research and there are lots of opportunities for reuse of old lessons and/or artifacts, for example.
J2EE consists of Servlets, JSP, JNDI, the new Deployment API, JMS, JCA, JTA, the JAX api's, and EJB. All of the new frameworks like Spring use these core API's to some degree.
You seem to only be complaining about EJB, and how it's going the CORBA way, which I find tremendously funny because EJB was originally (and still is) a standard runtime model for CORBA objects written in Java, with a difficult object/relational mapping spec bolted on.
That's the problem with Apple worshipers. They overlook Apple's problems to the point that they think Apple is perfect and Apple has no incentive to perfect their product.
I think this is a rather broad brush. And contradictory. In one sentence, Apple fans overlook problems in search for approval that Apple has the best "all around" product. In the next sentence, they're perfectionists.
why do people irrationally ignore the problems and attack anyone who is critical of an apple product?
If you were to look at fan-boys vs. detractors in terms of anxiety assumptions, Apple fans tend to have a big need for constant approval. My take: smaller market share, especially in the early to mid 1990's, led to a trend of knee-jerk attacks against anything "Apple" as bad / weird. This was prevalent in IT departments, and still is to some degree. People had their preferred platform, and felt sneered at.
I've seen this trend with OS/2 fans, I've seen it with Linux fans -- anyone who attaches emotional worth to their technology purchases tends to get anxious about the market share.
Now, this probably wasn't actually happening on a grand scale, there are a few vocal detractors in any community, and that probably just stoked a shared anxiety.
Now, Apple detractors (often alpha geeks, free software worshippers, or marketshare-obsessed capitalists) are that they exhibit perfectionist tendencies: they find one or two faults in the product or in the company that somehow makes the whole thing useless.
Now after a year of infrequent use, I get about 2-4 hours. This doesn't last me even half my day. I would NOT have bought the ipod mini if I knew this - competing similar products advertised battery lives of close to 20 hours.
Of course, these other products have no better battery life expectancy, they'll degrade all the same. Perhaps it will be more useful because they started at a higher level. But there is often a trade-off: the Dell DJ for example has much higher battery life but is VERY heavy. And then there the other features / bugs. For example, I have a friend that is pissed he didn't buy an iPod because his DJ locks up on him all the time, he has to carry a paper clip to reset it.
Anyhow, back to the battery issue. All batteries degrade: watches, cordless phones, mobile phones, laptops, PDAs, etc. There is a market for replacements. Some require professional service (watches?!). First gen iPods had worse batteries, but I wonder how many other hard drive MP3 players from 2001 are doing , battery-wise.
FWIW, my 1st-gen iPod from Autumn 2001 is still in use , with the original battery, and has just under 2 hours of battery life, and is used daily (I gave it to a friend, it fits within his commute). I use my 1st-gen iPod mini fairly regularly since Spring 2004, and its original battery life (8 hours?) is down to about 6 hours.
we see CASE tool technicians masquerading as software engineers.
I can guarantee you that a talented.NET or Java programmer abhors CASE tools as much as a UNIX guy.
I see this at work all the time with the "new generation" of IS/IT types rolling through the door who couldn't code a b-tree save to save their lives and rely on prebuilt everythings to give the illusion that they do something "difficult".
Why would I build something from scratch , so far removed from my real intent (to build an order management system)?
To counter your experience, I've seen plenty of UNIX folks that wrote server programs system that leaked memory and crashed often, and didn't really perform that well, all in the name of "it's better if I write it my way". I"ve also seen plenty of Java folks do the same, though this time in the name of "I can't be bothered to research what I'm re-using, whether a library or an RDBMS". Ignorance has many faces.
Natalie Portman has been a fixture topic on this site since 1999, when she first starred in Star Wars.
I think bitching about topics like this and the Bruce Campbell interview show a profound disrespect for the quirky nature of the topics here and the history behind why some topics come up repeatedly.
I come to Slashdot because I tend to enjoy these topics along side my dose of OSS , Linux, or whatever tech news. I would be upset if the editors listened to your voices and stopped posting articles like this , or the Bruce Campbell, interview.
Well, I find some of the poster personalities fascinating and entertaining. And I find the quirkiness of the editors (Taco in particular) amusing.
But mostly I'm still here because I see the odd Insightful post that turns me onto something new - an idea, a project, etc. Or I get into a good debate about some kind of subject.
For example, I've known about ML, Caml, OCaml, etc., but I've frankly never heard of SML/NJ. So I'm checking it out. I deal mostly with mainstream stuff (working for BEA Systems), but I do like to keep tabs on the future.
As for "mob over man", I think you're quite right that all communities devolve to this, and Slashdot always had this (remember the KDE/Gnome flamefests?), which is why karma and moderation was introduced. I think it works fairly well compared to other large scale message boards.
Slashdot.... It's messy, annoying, dirty, occaisionally funny and rarely, but somewhat continuously, insightful. It's just disappointing that the new class of "nerd" seems to be more interested in narrow topics over broad ones.
I mean, even in the "good 'ol days" the KDE vs. Gnome wars were incessant.
Saying "XYZ sucks!" I guess is a bad example. What I really am complaining about are the "What is XYZ? How is this related to Linux? How is this news for nerds?" posts.
Chances are there will eventually be more legitimate traffic than illegitimate. Now, I'm not saying the illegitimate market will shrink -- I'm saying that online content will grow, and legit means of content distro will eventually grow faster than pirate means.
1. Co-worker had a flaky iBook G3 in 2001, under warranty. Apple support shipped her a new one, worked fine. Another co-worker had a DVD drive issue in their Powerbook G4/500, Apple fixed it for free, no problems since.
2. My Powerbook G3/400 ("Lombard") -- heavy use from 1999-2001, no issues.
3. My first generation Powerbook G4/500 ("Titanium") -- heavy use from 2001-2003, no out of the box issues. I dropped it once, but didn't cause any problems other than warping the side of my case. Before reselling it, I had to get a key replaced on the keyboard and replace the bottom titanium plate, and purchase a new battery. In total, that cost me maybe $500 CAD.
4. My Powerbook G4 17-inch 1.0 GHz, which I owned from May 2003 through November 2003, was flawless. I upgraded to a 1.33 GHz, which I'm typing this post on, and it also has been great. I've had to replace my power supply once (due to the side-mount power plug, the wire bends a lot when I'm on the couch in the living room -- it eventually frayed). I may get a new battery since I've been using it on the road a lot and it's starting to lose its juice.
As an individual, I've been pretty happy with Apple hardware.
I will admit the only time I've been "unhappy" with Mac hardware was when I worked in an office of Performas that would hard-crash regularly on Mac OS 8 -- this was back in 1996.
One of the reasons I joined Slashdot in the first place (in late 1997) was the wide variety of geek-like interests that CmdrTaco and Hemos held -- from Linux / BSD / UNIX, to Perl, free sw / open source, Python, C, or Java programming, to gaming, to crackers/pirates, to Jon Katz's sometimes-interesting rants on culture, to cult movies, to casemods, to online rights, and a curious & growing interest in Apple (Jobs had just rejoined), etc.
But now there seems to be a lot of geeks that have a real track mind. They don't know classic cult TV or movie heros. They bitch about music / musicians because they're not a kernel mod. They hate TV or movie news (Battlestar sucks, Firefly sucks, Buffy/Angel sucked, Babylon 5 sucked, etc. --- What do you watch, The Weather Channel?).
Basically, all they want to talk shop! It's all about is Linux, or FOSS license pro/cons, or how Apache, MySQL, perl, etc. will somehow create a New Geek World Order.
What happened to the renaissance geek? Where's the passion for the obscure and beautiful, no matter the subject area?
It's my opinion that a CPU-intensive Java application on Intel or AMD will almost always smoke SPARC by a major factor. (I work for BEA, but I don't speak for them.)
For thread/throughput intensive applications, SPARC may win vs. P4 or Xeon (especially the multi-core SPARC models), but against Opteron or Itanium it's probably going to be a wash.
Generally we look at SPEC benchmarks when capacity planning, there is no inherent CPU advantage on the Sun JVM on Intel vs. Solaris (though the Solaris VM does have extra memory options like using Intimate Shared Memory which can be a benefit for large heap garbage collection).
And then there's our JRockit VM which only runs on Intel and it's been known to be extremely fast... though we're eventually going to port it to other chip architectures.
I've built designed databases and/or application server farms for coporations in telecom and finance.
If you need 64-bit addressing (i.e. large shared memory sizes, or individual process heap sizes over 2GB), you basically have 4 choices: Opteron, SPARC, POWER5, and Itanium2.
If you take a quick look at the SPECcpu2000, SPARC is a condenter if you need throughput, but not if you need raw CPU power. POWER5, Opteron and Itanium2 are about on-par, with minor (~10%) differentials between each other, in both power and throughput.
So really, it comes down to what hardware vendor do you want to do business with, and what software support is on each chip/OS, and that determines the chip you'll buy. Most corps deal with the big 3 (IBM, HP, Sun). HP chose Itanium2, Sun chose SPARC or Opteron, and IBM chose POWER5. Some systems are only tier-2 support on AIX POWER5, for example, but are tier-1 on Itanium2 HP-UX. Some are the other way around. Go figure...
HP has no where to go *but* Itanium for its small-SMP, Superdome and NonStop lines. That article noted they're investing $3 billion in using that chip over the next few years.
And Microsoft does support Windows Server 2003 on Itanium, that article was referring to client XP (which makes sense -- there is little/no need for an Itanium workstation at this point.)
I can get HP-UX, Linux, or Windows Server 2003 on HP's Itanium2 machines (I've had several running HP-UX and Windows in a corporate lab).
It strikes me that people use non-words as a way of hinting a direction to those who want change, and providing a form of plausible deniability to those forces that are against change.
This isn't restricted to business either; arguably government (civil service or military) is worse! Power politics exist in all human organizations and the basic dynamics have not changed in centuries. Even democratic systems don't change this base level of politics. Perhaps all that's changed are the value systems that drive courtship: from gaining favour with the sovereign to gaining favour with your boss and worshipping a narrow view of economic theory...
Take a look at IBM. They're moving to a services-based model.
Do you really want a world where software is maintained by IBM Global Services (and managed by former members of PriceWaterhouse Coopers)?
Services and consulting gigs are big bucks today, and getting bigger. Millions of dollars, usually. High hourly rates. And busloads of consultants of varying quality.
The services-based model isn't all it's cracked up to be, and I'm not sure it's the most appropriate way to ensure "freedom of code".
Who cares, honestly--and more to the point, who actually uses those RSI inducing apple mice for more than a few days? Grandmas, and people who want their mouse to look like their Apple. Think about it... It's true!
I'm a software developer. I have an Apple Bluetooth mouse that I use constantly with my Mac. My wrist is very comfortable and straight with it. How is it RSI inducing, or are you just being facetious?
Anyway, I'm usually typing at the keyboard so if I need to get a context menu I have my pinky on the Control key.... It's become a habit enough that it's second nature.
I don't see the point here, it's true for many IT companies, wether successfull or not.
My point was that very few companies have the level of talent that Google has attracted, and is paying for. There is a scarcity of such talent. Thus it's unlikely that Google's approach is scalable to beyond a handful of companies.
Progressing art & science in a market system usually implies innovation, and innovation usually implies profit. Profit isn't necessarily a reward, though it could be used as such. Profit's function in an economic system is covering the costs & risks of future development.
Limit all copyright times to the minimum required to pay back for creation costs (along the lines of 5 years).
Limiting terms is fine, and the current trend for unlimited extensions is dangerous, but I disagree that it's about covering creation costs. It's about creating a market for content, thus ensuring revenue flow for the creation of future works.
Cancel copyright on functional information (such as software). The power it grants the copyright holder over its user, even in a limited time, is too great.
I'm curious why you would think this. Copyright is what allows things like the GPL to exist. Without it, you don't have a community of open source with forced contributions, you have public domain artifacts.
Software creation, in most cases, requires little to no financial incentive
In most cases? In general, this could be applicable to any profession in which one gains pride and/or fellowship from their work -- Habitat for Humanity building houses, or Amish barn raisings at one end of the spectrum, pro-bono legal work as another example.
Just because financial needs aren't the ONLY incentive, this does not eliminate the fact that people need money.
and in niche cases where it does, payment to programmers is still possible.
Niche cases? Those niche cases would be where someone spends 8 hours a day developing software, and thus don't have time to make money in exchange for another form of labour? That's a strange definition of niche.
Let's break out this scenario....
Software creation, as with all forms of human activity, requires incentives. Financial incentives certainly aren't the only incentive. However, if one is to spend the majority of their time creating software, they require financial incentive. That means a wage, or a salary.
Wages and salaries must be paid by people or groups of people that undertake some kind of activity that provides economic value. Thus, they too must have incentive.
In a world where software licenses are no longer valued (i.e. public domain artifiacts), then the value is in: a) the time you spend (e.g. customization or support time); or b) the complementary products you associate with the software (e.g. retail websites, advertisments on the web, or selling hardware or business consulting) c) the usage of the software (e.g. software-as-a-service, metered usage, etc.)
So software-for-hire is developed by a consortium of volunteers in their spare time for certain classes of software plus full-time developers that are remunerated by manufacturers or software-service firms, or consulting / support firms.
Is this the model you seek? Is that really superior to today's model? I wonder.
Most popular open source software today is subsidised by hardware sales, business consulting, support contracts, and advertising (IBM, HP, RedHat, OSDN, Google, etc.).... Is this sustainable if the hardware business starts to falter, or if the business consultants lose large deals?
I do agree something needs to be done about the perpetual tax placed on desktop software upgrades, but I think that's slowly fixing itself -- people are upgrading less as the software becomes more commoditized and clones/alternatives appear. It's a long process, but probably in the next 10 years, Office won't be the cash cow it is today for Microsoft.
Allow copyright, but only apply it to inter-legal-entities copying. This would mean that EULA's have no effect (You really shouldn't need extra permission from the copyright owner to run the copy you bought!).
Seriously... what does the image of a company's CEO have to do with its revenue and profitability? Do you think Microsoft got to be where it is because of the mystique of Bill Gates? Perhaps it was a factor, but I don't think it was the primary one.
Certainly in the TV era there is a need to be able to be telegenic, a good presenter and speaker. Some CEO's are none of these things. Other than the odd CNBC and Bloomberg interview, many CEO's of multi-billion dollar companies are rarely heard from in public.
IBM had Thomas Watson Sr., and then Thomas Watson Jr., who became reasonably famous people. The arrival of Lou Gerstner started their switch to services and rebound.
Actually, amusing anecdote, I remember the Comdex 1994 keynotes. Gerstner spoke about the "network-centric future", about how OS/2 was all internet enabled and everything would be connected, the whole economy would be effected in 10 years time. Gates spoke about the "paperless office". Gates' vision is still a pipe dream.... Gerstner on the other hand nailed it (too bad IBM didn't execute on it well enough).
Slashdot Mirror
Object to relational layers are useful for certain classes of problems, but can be easily misused, leading to horridly performing applications. YMMV.
I just hope more rigour and discipline can be instilled to get past the bullshit surface-level debates (how much documentation, XP = hacking?, pair programming is/isn't stupid) that really distract from the whole point of the exercise.
I would highly suggest the following to get a better understanding of how this might fit into a software context:
Toyota doesn't have a meticulously defined development methodology. Nor does it use PERT or Gantt charts to track its car development projects. Nor does it have a "project manager" to sequence & assign tasks (it relies on a "Chief Engineer" to coordinate the whole effort). Car development certainly involves a lot of written documentation, but there is no massive "requirements" document that is written up front -- there are lots of alternative requirements and design documents for sub-systems and parts and various tradeoff curves. The scope and requirements vary through the cycle of the project. There is also iteration on design, and lots of reuse of prior designs and artifacts.
Toyota designs cars in a way that many people in the Agile community would love to design software -- in an interative and incremental fashion, without the massive soul-deadening top-down process enforcement. And they deliver the best quality, time to market, and profits in the auto industry.
Lean product development is about iteration, and it's how Toyota & Lexus designs its cars, noted as the most reliable in the world. Why can't these techniques also apply to software, which is arguably even easier to change iteratively than car part designs?
And Toyota is slowly driving their competitors into the ground. I'd take another look. There's a lot more to Agile than the surface-level debates, especially if you look at it in the context of Lean product development techniques.
Six sigma is more about production processes, but Lean product development certainly has a lot in common with agile methods. It takes a bit more rigourous approach to the process though -- development certainly isn't research and there are lots of opportunities for reuse of old lessons and/or artifacts, for example.
J2EE consists of Servlets, JSP, JNDI, the new Deployment API, JMS, JCA, JTA, the JAX api's, and EJB. All of the new frameworks like Spring use these core API's to some degree.
You seem to only be complaining about EJB, and how it's going the CORBA way, which I find tremendously funny because EJB was originally (and still is) a standard runtime model for CORBA objects written in Java, with a difficult object/relational mapping spec bolted on.
That's the problem with Apple worshipers. They overlook Apple's problems to the point that they think Apple is perfect and Apple has no incentive to perfect their product.
I think this is a rather broad brush. And contradictory. In one sentence, Apple fans overlook problems in search for approval that Apple has the best "all around" product. In the next sentence, they're perfectionists.
why do people irrationally ignore the problems and attack anyone who is critical of an apple product?
If you were to look at fan-boys vs. detractors in terms of anxiety assumptions, Apple fans tend to have a big need for constant approval. My take: smaller market share, especially in the early to mid 1990's, led to a trend of knee-jerk attacks against anything "Apple" as bad / weird. This was prevalent in IT departments, and still is to some degree. People had their preferred platform, and felt sneered at.
I've seen this trend with OS/2 fans, I've seen it with Linux fans -- anyone who attaches emotional worth to their technology purchases tends to get anxious about the market share.
Now, this probably wasn't actually happening on a grand scale, there are a few vocal detractors in any community, and that probably just stoked a shared anxiety.
Now, Apple detractors (often alpha geeks, free software worshippers, or marketshare-obsessed capitalists) are that they exhibit perfectionist tendencies: they find one or two faults in the product or in the company that somehow makes the whole thing useless.
Now after a year of infrequent use, I get about 2-4 hours. This doesn't last me even half my day. I would NOT have bought the ipod mini if I knew this - competing similar products advertised battery lives of close to 20 hours.
Of course, these other products have no better battery life expectancy, they'll degrade all the same. Perhaps it will be more useful because they started at a higher level. But there is often a trade-off: the Dell DJ for example has much higher battery life but is VERY heavy. And then there the other features / bugs. For example, I have a friend that is pissed he didn't buy an iPod because his DJ locks up on him all the time, he has to carry a paper clip to reset it.
Anyhow, back to the battery issue. All batteries degrade: watches, cordless phones, mobile phones, laptops, PDAs, etc. There is a market for replacements. Some require professional service (watches?!). First gen iPods had worse batteries, but I wonder how many other hard drive MP3 players from 2001 are doing , battery-wise.
FWIW, my 1st-gen iPod from Autumn 2001 is still in use , with the original battery, and has just under 2 hours of battery life, and is used daily (I gave it to a friend, it fits within his commute). I use my 1st-gen iPod mini fairly regularly since Spring 2004, and its original battery life (8 hours?) is down to about 6 hours.
we see CASE tool technicians masquerading as software engineers.
.NET or Java programmer abhors CASE tools as much as a UNIX guy.
I can guarantee you that a talented
I see this at work all the time with the "new generation" of IS/IT types rolling through the door who couldn't code a b-tree save to save their lives and rely on prebuilt everythings to give the illusion that they do something "difficult".
Why would I build something from scratch , so far removed from my real intent (to build an order management system)?
To counter your experience, I've seen plenty of UNIX folks that wrote server programs system that leaked memory and crashed often, and didn't really perform that well, all in the name of "it's better if I write it my way". I"ve also seen plenty of Java folks do the same, though this time in the name of "I can't be bothered to research what I'm re-using, whether a library or an RDBMS". Ignorance has many faces.
Read up on your history, asshole.
This has been stuff that matters since around 1999. And by "stuff that matters", it generally is up to the editors, not the "majority".
Natalie Portman has been a fixture topic on this site since 1999, when she first starred in Star Wars.
I think bitching about topics like this and the Bruce Campbell interview show a profound disrespect for the quirky nature of the topics here and the history behind why some topics come up repeatedly.
I come to Slashdot because I tend to enjoy these topics along side my dose of OSS , Linux, or whatever tech news. I would be upset if the editors listened to your voices and stopped posting articles like this , or the Bruce Campbell, interview.
the scenery is better :-)
Well, I find some of the poster personalities fascinating and entertaining. And I find the quirkiness of the editors (Taco in particular) amusing.
But mostly I'm still here because I see the odd Insightful post that turns me onto something new - an idea, a project, etc. Or I get into a good debate about some kind of subject.
For example, I've known about ML, Caml, OCaml, etc., but I've frankly never heard of SML/NJ. So I'm checking it out. I deal mostly with mainstream stuff (working for BEA Systems), but I do like to keep tabs on the future.
As for "mob over man", I think you're quite right that all communities devolve to this, and Slashdot always had this (remember the KDE/Gnome flamefests?), which is why karma and moderation was introduced. I think it works fairly well compared to other large scale message boards.
Slashdot.... It's messy, annoying, dirty, occaisionally funny and rarely, but somewhat continuously, insightful. It's just disappointing that the new class of "nerd" seems to be more interested in narrow topics over broad ones.
I mean, even in the "good 'ol days" the KDE vs. Gnome wars were incessant.
Saying "XYZ sucks!" I guess is a bad example. What I really am complaining about are the "What is XYZ? How is this related to Linux? How is this news for nerds?" posts.
iTunes is more popular than LimeWire.
iStudy: iTunes more popular than many P2P sites
Chances are there will eventually be more legitimate traffic than illegitimate. Now, I'm not saying the illegitimate market will shrink -- I'm saying that online content will grow, and legit means of content distro will eventually grow faster than pirate means.
1. Co-worker had a flaky iBook G3 in 2001, under warranty. Apple support shipped her a new one, worked fine. Another co-worker had a DVD drive issue in their Powerbook G4 /500, Apple fixed it for free, no problems since.
2. My Powerbook G3/400 ("Lombard") -- heavy use from 1999-2001, no issues.
3. My first generation Powerbook G4/500 ("Titanium") -- heavy use from 2001-2003, no out of the box issues. I dropped it once, but didn't cause any problems other than warping the side of my case. Before reselling it, I had to get a key replaced on the keyboard and replace the bottom titanium plate, and purchase a new battery. In total, that cost me maybe $500 CAD.
4. My Powerbook G4 17-inch 1.0 GHz, which I owned from May 2003 through November 2003, was flawless. I upgraded to a 1.33 GHz, which I'm typing this post on, and it also has been great. I've had to replace my power supply once (due to the side-mount power plug, the wire bends a lot when I'm on the couch in the living room -- it eventually frayed). I may get a new battery since I've been using it on the road a lot and it's starting to lose its juice.
As an individual, I've been pretty happy with Apple hardware.
I will admit the only time I've been "unhappy" with Mac hardware was when I worked in an office of Performas that would hard-crash regularly on Mac OS 8 -- this was back in 1996.
I'm noticing a disturbing trend here.
One of the reasons I joined Slashdot in the first place (in late 1997) was the wide variety of geek-like interests that CmdrTaco and Hemos held -- from Linux / BSD / UNIX, to Perl, free sw / open source, Python, C, or Java programming, to gaming, to crackers/pirates, to Jon Katz's sometimes-interesting rants on culture, to cult movies, to casemods, to online rights, and a curious & growing interest in Apple (Jobs had just rejoined), etc.
But now there seems to be a lot of geeks that have a real track mind. They don't know classic cult TV or movie heros. They bitch about music / musicians because they're not a kernel mod. They hate TV or movie news (Battlestar sucks, Firefly sucks, Buffy/Angel sucked, Babylon 5 sucked, etc. --- What do you watch, The Weather Channel?).
Basically, all they want to talk shop! It's all about is Linux, or FOSS license pro/cons, or how Apache, MySQL, perl, etc. will somehow create a New Geek World Order.
What happened to the renaissance geek? Where's the passion for the obscure and beautiful, no matter the subject area?
It's my opinion that a CPU-intensive Java application on Intel or AMD will almost always smoke SPARC by a major factor. (I work for BEA, but I don't speak for them.)
For thread/throughput intensive applications, SPARC may win vs. P4 or Xeon (especially the multi-core SPARC models), but against Opteron or Itanium it's probably going to be a wash.
Generally we look at SPEC benchmarks when capacity planning, there is no inherent CPU advantage on the Sun JVM on Intel vs. Solaris (though the Solaris VM does have extra memory options like using Intimate Shared Memory which can be a benefit for large heap garbage collection).
And then there's our JRockit VM which only runs on Intel and it's been known to be extremely fast... though we're eventually going to port it to other chip architectures.
I've built designed databases and/or application server farms for coporations in telecom and finance.
If you need 64-bit addressing (i.e. large shared memory sizes, or individual process heap sizes over 2GB), you basically have 4 choices: Opteron, SPARC, POWER5, and Itanium2.
If you take a quick look at the SPECcpu2000, SPARC is a condenter if you need throughput, but not if you need raw CPU power. POWER5, Opteron and Itanium2 are about on-par, with minor (~10%) differentials between each other, in both power and throughput.
So really, it comes down to what hardware vendor do you want to do business with, and what software support is on each chip/OS, and that determines the chip you'll buy. Most corps deal with the big 3 (IBM, HP, Sun). HP chose Itanium2, Sun chose SPARC or Opteron, and IBM chose POWER5. Some systems are only tier-2 support on AIX POWER5, for example, but are tier-1 on Itanium2 HP-UX. Some are the other way around. Go figure...
HP has no where to go *but* Itanium for its small-SMP, Superdome and NonStop lines. That article noted they're investing $3 billion in using that chip over the next few years.
And Microsoft does support Windows Server 2003 on Itanium, that article was referring to client XP (which makes sense -- there is little/no need for an Itanium workstation at this point.)
I can get HP-UX, Linux, or Windows Server 2003 on HP's Itanium2 machines (I've had several running HP-UX and Windows in a corporate lab).
It strikes me that people use non-words as a way of hinting a direction to those who want change, and providing a form of plausible deniability to those forces that are against change.
This isn't restricted to business either; arguably government (civil service or military) is worse! Power politics exist in all human organizations and the basic dynamics have not changed in centuries. Even democratic systems don't change this base level of politics. Perhaps all that's changed are the value systems that drive courtship: from gaining favour with the sovereign to gaining favour with your boss and worshipping a narrow view of economic theory...
Take a look at IBM. They're moving to a services-based model.
Do you really want a world where software is maintained by IBM Global Services (and managed by former members of PriceWaterhouse Coopers)?
Services and consulting gigs are big bucks today, and getting bigger. Millions of dollars, usually. High hourly rates. And busloads of consultants of varying quality.
The services-based model isn't all it's cracked up to be, and I'm not sure it's the most appropriate way to ensure "freedom of code".
Who cares, honestly--and more to the point, who actually uses those RSI inducing apple mice for more than a few days? Grandmas, and people who want their mouse to look like their Apple. Think about it... It's true!
I'm a software developer. I have an Apple Bluetooth mouse that I use constantly with my Mac. My wrist is very comfortable and straight with it. How is it RSI inducing, or are you just being facetious?
Anyway, I'm usually typing at the keyboard so if I need to get a context menu I have my pinky on the Control key.... It's become a habit enough that it's second nature.
I don't see the point here, it's true for many IT companies, wether successfull or not.
My point was that very few companies have the level of talent that Google has attracted, and is paying for. There is a scarcity of such talent. Thus it's unlikely that Google's approach is scalable to beyond a handful of companies.
And no, folks, it is not meant to reward authors.
Progressing art & science in a market system usually implies innovation, and innovation usually implies profit. Profit isn't necessarily a reward, though it could be used as such. Profit's function in an economic system is covering the costs & risks of future development.
Limit all copyright times to the minimum required to pay back for creation costs (along the lines of 5 years).
Limiting terms is fine, and the current trend for unlimited extensions is dangerous, but I disagree that it's about covering creation costs. It's about creating a market for content, thus ensuring revenue flow for the creation of future works.
Cancel copyright on functional information (such as software). The power it grants the copyright holder over its user, even in a limited time, is too great.
I'm curious why you would think this. Copyright is what allows things like the GPL to exist. Without it, you don't have a community of open source with forced contributions, you have public domain artifacts.
Software creation, in most cases, requires little to no financial incentive
In most cases? In general, this could be applicable to any profession in which one gains pride and/or fellowship from their work -- Habitat for Humanity building houses, or Amish barn raisings at one end of the spectrum, pro-bono legal work as another example.
Just because financial needs aren't the ONLY incentive, this does not eliminate the fact that people need money.
and in niche cases where it does, payment to programmers is still possible.
Niche cases? Those niche cases would be where someone spends 8 hours a day developing software, and thus don't have time to make money in exchange for another form of labour? That's a strange definition of niche.
Let's break out this scenario....
Software creation, as with all forms of human activity, requires incentives. Financial incentives certainly aren't the only incentive. However, if one is to spend the majority of their time creating software, they require financial incentive. That means a wage, or a salary.
Wages and salaries must be paid by people or groups of people that undertake some kind of activity that provides economic value. Thus, they too must have incentive.
In a world where software licenses are no longer valued (i.e. public domain artifiacts), then the value is in:
a) the time you spend (e.g. customization or support time); or
b) the complementary products you associate with the software (e.g. retail websites, advertisments on the web, or selling hardware or business consulting)
c) the usage of the software (e.g. software-as-a-service, metered usage, etc.)
So software-for-hire is developed by a consortium of volunteers in their spare time for certain classes of software plus full-time developers that are remunerated by manufacturers or software-service firms, or consulting / support firms.
Is this the model you seek? Is that really superior to today's model? I wonder.
Most popular open source software today is subsidised by hardware sales, business consulting, support contracts, and advertising (IBM, HP, RedHat, OSDN, Google, etc.).... Is this sustainable if the hardware business starts to falter, or if the business consultants lose large deals?
I do agree something needs to be done about the perpetual tax placed on desktop software upgrades, but I think that's slowly fixing itself -- people are upgrading less as the software becomes more commoditized and clones/alternatives appear. It's a long process, but probably in the next 10 years, Office won't be the cash cow it is today for Microsoft.
Allow copyright, but only apply it to inter-legal-entities copying. This would mean that EULA's have no effect (You really shouldn't need extra permission from the copyright owner to run the copy you bought!).
Hm
Seriously... what does the image of a company's CEO have to do with its revenue and profitability? Do you think Microsoft got to be where it is because of the mystique of Bill Gates? Perhaps it was a factor, but I don't think it was the primary one.
Certainly in the TV era there is a need to be able to be telegenic, a good presenter and speaker. Some CEO's are none of these things. Other than the odd CNBC and Bloomberg interview, many CEO's of multi-billion dollar companies are rarely heard from in public.
IBM had Thomas Watson Sr., and then Thomas Watson Jr., who became reasonably famous people. The arrival of Lou Gerstner started their switch to services and rebound.
Actually, amusing anecdote, I remember the Comdex 1994 keynotes. Gerstner spoke about the "network-centric future", about how OS/2 was all internet enabled and everything would be connected, the whole economy would be effected in 10 years time. Gates spoke about the "paperless office". Gates' vision is still a pipe dream.... Gerstner on the other hand nailed it (too bad IBM didn't execute on it well enough).