I agree. Likewise should the US government forbid acquisition of any tools, machinery, parts, components or other tangible or intangible entities which might conceivably hide some kind of intelligence intercept mechanism if said items are not manufactured in full within the borders of the United States of America. It's simply not worth the risk.
I agree with you to an extent. However in my experience (having worked as a consultant for 10 years with mainly larger companies and a few smaller in various countries) I'd say the the platform adoption varies drastically between countries in general as well as for different markets (inside as well as outside a countries border to make things more complicated).
The highest Microsoft penetration per geographic area I've seen has been the Nordic region of northern Europe. The lowest in south america. For each country though, the pattern has been that larger companies are mostly using non-Microsoft platforms in the server park (Even in the Nordics except for their government organizations).
Now, in my experience there are two primary differences between larger and smaller companies when it comes to vendor selection and acquisition of vendor products.
First off, big companies usually have rigorous standardization in place and IT policies which often dictate what exact product should be used for what type of service in the enterprise. This means that a new vendor has a real uphill battle to sell products to an enterprise if that product is solving the same or similar problems as an already deployed one (from another vendor). The reward on the other hand if the vendor manages to convince the enterprise to switch is enormous since the policy will likely be changed to mandate the new product for future acquisitions and deployments instead of the old one.
Secondly, enterprise customers are very picky and have a stack of pages listing functional, non-functional requirements page up and down, many of which will not be matched by any vendor's product. This requires the vendor to adapt the product (by itself, together with partners or less often together with the customer's IT-staff) in order for it to pass the check-list criteria. Now in order to get the advantages of the previous paragraph, the vendor must be able to pass this checklist and given that very few vendor packages out-of-the-box(es) I've seen have been able to meet all checkpoints of any major client's list, there is bound to be need for custom software adaptations. AFAIK, Microsoft's business model is not to adapt their products to specific client needs and neither to build them in a way which allows them to be easilly adapted, integrated with non-MS products (of which most will be in a given enterprise). Companies like IBM on the other hand knows this game very well and make sure they can integrate with pretty much any other vendor's product they'll encounter in the client's server park. IBM which is also mostly a service (consulting) company has the manpower and skills needed to do whatever adaptations are needed to their offerings.
As I see it, Microsoft has a real challenge before them. In my view, they must also become a service company in order to compete with the level of service enterprises have learnt to expect during the past decades. However I'm not really sure this is aligned with the future vision of the company as Mr. Gates sees it, which might prove to be a dilemma as well as a non-insignificant hurdle to overcome.
What I was referring to was the implementation, not the ideas governing the implementation. Take two different implementations of the same software idea and you'll surely see two different implementations. This even stretches as far as to implementations of the same idea at different points in time by the same developer.
Don't mend what is not broken rarely applies to software developers (unfortunately). Component reuse was a buzz word a few years ago which unfortunately didn't have much of an impact for inhouse development in cross project form. Developers like to experiment with new stuff and applying new algorithms or patterns to a known problem seems to keep them happy (avoids them being turned into factory workers i guess) and gives them a way to further educate themselves. However, this often impacts the deliverables in a (from a product manager's perspective) negative way since it introduces more unknown factors and deviations with each re-implementaion of a problem solution, than would seem necessary from the inception outlook.
Engineering in other segments (like the tangible sectors) of the market is simply more strict than in the software market, the latter where people build castles from thin air (pure thought stuff). In software the cost of trying out new approaches to problems is significantly lower than in the tangible markets which gives bored programmers a whole new set of freedoms when it comes to experimentation.
On a macro level though, the same governing rules apply as in the tangible segments. Stick with a proven formula since it minimizes risk. Yeah, it's boring and yields unimaginative products but that's what you get when the Harvard grads are sitting on the money bag.
The point I was originally trying to make was that when you pre-order stuff, you do it when you pretty much know what the product will be like. Final products can be predicted pretty well in the utility market but not in the entertainment segment. Utilities are built, marketed and sold upon proven formulas and solid history while entertainment are (or should strive for) about providing new experiences which may often require unproven formulas (higher risk and subjective in appreciation which if appreciated by many can bring in the jackpot). Since entertainment is based on subjective experiences it lends itself poorly at pre-ordering strategies from a consumer's point of view (the unpredicatable nature of the result that is).
Because it's better than simply reading the title and a summary at Amazon.com ?
Really, what's your point? People pre-order cars since most cars are just new revisions / bugfixes to older models with very little changing over each revision (such as the yearly increments of the BMW E46 model for example). I don't think the car business and their merchandise can be compared to the software industry and theirs. Programmers prefer to re-invent the wheel far more often than any other engineering profession.
The gaming industry is a segment of the entertainment industry and as such the same rules governing other practical / utility-industries do not apply. If I pre-order a non-software utility gadget which builds on an existing model (which is often the case), I know pretty much what I'll get. With entertainment this simply isn't true. Thus and apples and oranges comparison.
Today most laptops are just dragged between electric outlets since they wont go for more than a couple of hours without recharging
Agreed. Laptops are mostly used as portable laptops from my experience as well. This however is not affecting productivity as far as I've seen. At work we typically roam from room to room with our AC-adapters for meetings, presentations or getting some work done in the labs. Coupled with this we also roam to different client locations which all have power outlets for our adapters. Point is that the pattern is people moving from location A to location B and spending the majority of time not moving around. When stationary, plugging in the AC-adapter is not a big deal.
However, compare this to a full blown stationary setup with a 15Kg case and a 7Kg monitor, not to mention all the associated gadgets such as keyboards, mice and cables, the true benefit of a laptop for a person who isn't completely stationary at one site becomes clear. A all-in-one computer to bring with you to each new location.
Battery really isn't that big of a problem from what i've gathered from clients in most industries. What people in my experience clamor for is the same stuff as desktop users want; More speed and, better displays and more storage.
Sure the ongoing 10% yearly increase in battery life is nice, but other attributes take precedence.
PS. Imagine the new logistic problem with fuel cells for companies with multiple sites. "6 hours" (9 hours and subtracting the 1/3 optimistic marketing figure addition) on each kit, meaning I'd use about 2 a day. At our office alone, we'd have to stock up with 8000 each week. The company as a whole would burn through about 1.6 million a day or about 350 million kits/refills a year... For one company alone.
A thought just crossed my mind. Since admins running Unix-like systems mostly operate as non-root users, wouldn't it then be possible for a malware to lurk in the background of the non-privileged sandbox until you sudo/su and then for it to use the newly gained privileges to wreak havoc/gather intel and hide itself? In a non-root sandbox the malware process would likely show up in the process list, but who can honestly say that they check the process list each time before they become root? Also, a malware naming itself like a common process (or the same as a process which already occupies the list) like bash for example, would make a casual glance likely to disregard that listed entry.
Disclaimer: I don't pretend to know the intrisinct details regarding privilege escalation in *nix, so this thought might well be nothing but nonsense.
First off, I think you totally missed the point regarding what Trusted Computing really is all about. Giving ultimate authority over stuff you buy to the vendor instead of you (the user/consumer/prosumer/sucker buying the stuff).
Secondly, you talked about something else, namely information loss and that you claimed you could prevent it. Here your arguments are vague and rather lacking i.m.h.o. The only secure system is a system without external interfaces. Such a system is on the other hand not very practical, so essentially all system have them. Without knowing what you use your systems for or how, I will venture a guess that some of them at least are networked. Further, if you with these systems provide any kind of information to the outside world (e.g. web servers) then there are countless schemes to get information out of your machines without you having the simplest notion the occurrence.
Example, say you get hit by the "latest" Apache / IIS / Bind or WhateverApp virus, which happens to modify said app to do malicious information gathering. After having gathered the information it can then use your compromised application as a relay to allow the virus author to obtain the information (e.g. through some protocol like for instance inside a specific web page in base64 encoding, specially crafted ping replies or whatever). Point is, if someone is able to compromise your applications on a networked machine, then there are countless ways of getting that information out without you having the slighets clue.
Note on boasting; Hubris is what brought Icarus to his death after all.
Well, in Europe right now, there is a big conversion going on towards GNU/Linux. I've personally been involved with the switch of three government organizations in two countries on the server side and one on the desktop. From colleagues in other countries (esp. Germany) they are succeeding to migrate to a better platform even faster.
As more people become accustomed to alternative platforms and applications, they tend to use them at home as well. Nothing substitutes learning like the massive hands on experience at work or in schools.
Would MS be forced to exit the EU, it would probably be under a grace period suitable for an accelerated program of rolling out alternative platforms and tools. Not over night, so I would not be worried.
Main problem still remains,the lack of transpareny
on
No Backdoor in Vista
·
· Score: 3, Insightful
The problem is transparency. Would you stake your business or for that matter, you life (as is the case in some regions of the world) on this assumption? Since there is no transparency in Microsoft products, you simply have to take their word for it.
I thought the golden rule of security was that any viable security mechanism should tolerate public scrutiny. Knowing how the software works should not work against the devised scheme itself.
If this isn't vapor ware, why in G*d's name do they alert the competition of this fact? I thought the normal practice was to slam your competition suddenly and without warning, which in my book doesn't include touting future plans before you're forced to do so.
Unfortunately, my modpoints are all out at present, or I would have rated your post insightful.
In my view, the US government is only listening to one voice, the voice of military offensive power. In that light, since it's proven mostly futile to reason with the US regarding important (as well as less important) subjects, perhaps the rest of the world should aim at restoring some balance of power. This is really the last thing I would like to see happen, but the US really doesn't seem to be giving the rest of the world much of a choice. Some ramp up in offensive power in the EU as well as China should hopefully counteract some of the imperialistic agression of today's United States of America and help draw some much needed lines.
Yes, treacherous computing will keep the digital roads safe and locked down. Pay or we exclude you from using your own computer.
All information will be "pay-per-view". Step one is to ensure nothing ever again sips out into the public domain. Step two is to protect the concept of every idea and make people pay per use for each. Instead of "thoughts are free" the saying will be "thoughts might be available for a pre-negotiated price". Without enough money, your child won't be implanted with the genes required for a concious mind. You'll get a vegetable.
We're heading straight towards this future. I really do hope people remember why the right to keep and bear arms was so important... Someone better come up with a good alternative idea (and fast!) to change the current horrid path.
"If you don't like it, don't buy it". This advice is becoming rather repetitive. It's also not an efficient way of "signaling" to a company that the XYZ they do sucks and they should stop.
If someone regularly (once a month?) spends around $100K or $1M on stuff from a given company and suddenly stops, then that might raise an eyebrow and perhaps trigger the company to question why the person stopped the regular spending of money. Anything less will *not* react much of attention.
On the other hand, a person whose opinion is voiced (blog, newspaper etc.) will generate a *much* larger impact. Especially since most larger companies have entire departments scouring the web and other media for tidbits about them. Also, by voicing their opinion, the company gets to know *what* they did wrong. Just stopping the money spending tells them little to nothing about what's wrong.
So, don't kid your self that a single person "not buying an i-pod" for $200 will result in any specific change at companies like Apple.
Perhaps this "fury" over voting on the "other guy" could be addressed by having more candidates? Now, I realize it's in the politics 101 book to always envision "an enemy", since it's a tactic for making people who would otherwise not agree come together. However, as I said, it's mostly a tactic by politicians *for* politicians and eliminating or weakening this tool would be a *Good Thing* in my book.
So, again, why not introduce more candidates? Also, why not count each single vote by each single voter cast in the election? Today's state lumped voting mechanism isn't really working in a democratic election when the candidates are from the people (any given citizen deciding to run) as a real democracy would allow.
Finally, the argument that "people are too stupid/lazy to keep track of more than two things at a time" doesn't fly with me. Even if you had 1000 candidates, there would be people who would not vote/care, same as it is in today's system trading whitches for devils. It doesn't mean a system with 1000 candidates is any less democratically efficient than one with 2 candidates. More candidates simply means higher probability of finding a more suitable candidate in whom you would entrust your next 4 years to. This would likely result in greater number of active voters since a suitable match would be more likely. I think one of the current problems is that neither of the current parties appeal to a lot of people and so people simply choose to not "get involved". This is no worse than would the situation be had these poeople arbitrarily picked a candidate at random for voting on, right?
More choices equals more freedom and allows for greater amount of expression.
I'm fortunate to have a 10MBit symmetrical connection from my ISP, soon to be beefed to 100Mbit. I also know that the ISP is shaping some P2P protocol traffic of which bittorrent is one. However this ISP realizes who it's customers are and that someone ordering 10MBit internet access is probably going to use it to some extent since they didn't go with another provider offering less capacity. So, the ISP has a pretty good infrastructure in place with good peer agreements with other ISPs. Now, even though they have a fairly well developed infrastructure they will run into extreme peaks and must be able to manage dataflows so that the main internet services are available to their customers (telnet, SMTP, VOIP, HTTP/S, SSH etc.). So they use traffic shaping to simply assign lower priority to P2P traffic.
This behavior is totally OK with me and I wouldn't have it any other way. Certain services are more time critical than others and I wouldn't like them to be affected by the huge P2P clogs in the network. On the other hand, since they just re-prioritize the packets I know that my P2P transfer rate will be good enough (the remaining capacity of the network when the essential services are cared for). This solution works great when you have an ISP which continually beefs up their backbone, but would pose a problem if you have a cheap *ss ISP which models their backbone capacity only for the (by them considered) essential services.
So, in short, if you have an ISP which do cater for your specific customer type, then traffic shaping can be a good thing for you as well as your fellow ISP customers. Traffic shaping isn't all evil, that's all.
I kind of understand what you are trying to convey. When I write a program for myself or in a team with just a few developers who know how every aspect of the solution works, it's nice to change stuff at a whim. Since the team is small, everyone monitors each cvs-update for changes and talks to each other frequently to keeps abreast of what has changed. However, for projects spanning more than just a handful people, specs quickly becomes essential as a communication tool.
An API is just that, a communication instrument (or protocol) which is intended to describe the exposed interfaces, their purpose, what they do when used in various ways etc. Essential information for using a component. This is information any user of the component would need to know anyway in order to use it properly.
Now, when you mentioned refactoring being sacrificed if interfaces (APIs) are published (which is what I interpreted your text to say), I completely failed to see the underlying reasoning.
Refactoring is a disciplined technique for restructuring an existing body of code, altering its internal structure without changing its external behavior.
(reference). What this seems to say is that the whole point of refactoring is to keep the code from decaying while not changing the interfaces!
If you require new features / behaviors, why not simply create new function signatures and deprecate old ones? What's so difficult or bad with that approach to changes?
Somehow I don't think that's really what the parent poster meant. ATI could very possibly provide passable support if they wanted to, but they don't. ATI is notorious for their total lack of interest in Linux. I agree with you, that any sane company aiming at establishing a foot hold in an emerging platform market would jump at the opportunity to make their own products work great on said platform. However ATI is not such a company. Sad as it might seem, ATI still doesn't see enough value for providing adequate product support on Linux. They're suffering from the hen and chicken syndrome, hoping that the effort spent on RedHat and SuSE (since that's what Linux basically is to them) by their competitor(s) might some day make it grow to a viable market segment for them.
... they will still have the excitement of being an early adopter...
Why would anyone be excited at being an early adopter? Isn't it common sense to let a product reach a certain level of maturity instead of playing the part of a ginnie pig / extended QA arm of a company, having to pay for the displeasure of finding and experiencing kinks in the offering?
Observe and adopt when the time is right I say... When the flawed fans have been replaced, the heating problem resolved, the availability of enough accessories to add proper value to the offering etc. You know, the typical scenario for new products.
They do. Ever looked at the credits after a movie? I'd wager 1/3 to 1/2 of the actors in the list have changed their name. Either because their previous names were too convoluted for the average American to pronounce (e.g. they might have ancestral ties to other countries) or they had too common names, meaning they needed to change them to stick out just enough. This example simply shows that names are very important to us when it comes to instances of types.
When it comes to the specific types (e.g. types of technology) themselves however, we tend not to care that much what names we use to classify the different things with. Since many of these things stem from science (discovery) we might lack prior art and have settled with the fact that "new stuff will yields new names". Our perception of the new stuff is formed by what it can do for us (after having seen or read about it) and so "strange" names like Turbo, Laser, Gasoline, Homo sapien etc. get established and recognized by many as types.
What would help tremendously though when naming new stuff (types) would be to look at the traits of the concept and pick a name communicating the intent of the type. This would help people quickly pick up what the thing is about and speed up awareness immensely.
E.g. Say you invented a strap which you could tie around a dog's neck and also attach a leach to. Now instead of giving that thing a name like Smorgasbord, wouldn't a name like Dog Collar be easier to remember? Wouldn't it also convey better the intent of the device to people not yet familiar with this new thing? The concept collar is already familiar for most people and to distinguish this new type variant from the one humans use, adding the type name of the intended wearer will avoid any confusion in the morning.
So, is Joomla a good name? Perhaps not. Will it be catastrophic? perhaps not. Could a better name have resulted in wider awareness in shorter time? Personally I think so.
If you manager produces noticeable value in the area for which s/he was hired (technical or non-technical) for you, your team and the rest of the company, then s/he is probably in the right spot.
Managing and being at the fore front of technology is VERY taxing and tends to result in burn outs of most people. I've seen people being able to cope with such jobs at a maximum of 18 months at a time before requiring downtime (e.g. focusing on either for a while) or they'd be hitting the wall (too many have gone down the latter road for my liking).
Personally, I've been in a position where I for the last 12 months have had to negotiate contracts, make planning, acquire resources, deliver architecture and manage the execution and delivery of systems for a government contract. If it were not for the fact that I'd be handing this over to a replacement in 1,5 months I'd burn myself out.
When I eventually do take a position in a line organization I'll definitely choose to *either* focus on sales, resource management, project execution or architecture. Not all of them at once since it's unrealistic to expect an "IT Manager" to keep "consultant" like pace as part of a long term career (5+ years?). No person I've ever met have ever managed such a feat and expecting such from your manager is not realistic i.m.h.o.
And yes, if when the time comes to choose a path and the path happens to be non-technical, I will become less skilled in the areas not being the core of my new position. E.g. I would not see myself being able to teach younger specialists (as an example) the ins and outs of various techs. I'll expect any person I hire or manage to be better than I in their respective field and will rely on their advice and expertise for making the proper decisions.
Please reflect over your manager's situation and the work s/he does. In most likelihood the person is busy making sure you and your team can perform to your best while at the same time "shielding you " from upper management and the issues coming from that direction. I'd also wager SLA , customer commitments and allocated budgets weigh heavy on your manager to name but a few examples.
At a talk RMS held in Germany a year ago or so, he mentioned that a *friend* of his had gone through the Linux source in an attempt to find possible patent infringements. RMS mentioned his friend found thousands, so regardless of Perens being a d*ck or not, there might be some substance to the claim.
Not everyone is hyped about the iTunes service. Therefore I hope Google will make it possible for users to include iTunes hits. Otherwise this deal results in instant degraded search results for a lot of users (those not interested in itunes).
Personally I despise the service for one major reason. The quality of the purchased sound SUCKS! To be fair and giving Apple *some* credit, the packaging (the application with it's simple user interface and it's preview functionality) is nice. However the wares being sold is of lousy quality and the primary point of iTunes was selling sound (at least what I though).
I also wonder if the EU has a case against Apple for errecting trade barriers. Not all songs are made available to all member countries of the EU which is another thing deeply disturbing to me (some songs for ex. available in the UK store are not available in the German).
This seems like an excellent book and subject. The industry is in dire need for more people to write for the *need* of other people.
Two current examples. My company employ one of the big development methods, aging back some 10-15 years. It contains templates for various (document) deliverables the company pitch as an advantage to customers. These documents (ranging from 80-500 depending on the project) are something each project spits out *just because* they have to (lousy project management). The decision of what document deliverables to produce is made something like 1-2 years ahead, which in reality means only 5-10% of them would be useful at the date of shipping. Instead of having all these tens of thousands of useless crap pages being manufactured, junk which no-one will ever read, it would be much more appropriate if the right people could write about the right thing in a way suitable for the intended audience. Unfortunately this would mean a break from the *method* at various point which seems unlikely at present given the particular company culture. In short, Agility needs to enter the "document manufacturing" cycles as well.
The second example has to do with with doers vs. explainers (is that a word?). I've run some projects in which we had some brilliant designers and coders, creating the most flexible and elegant designs and implementation. Unfortunately these people either didn't document them or documented them poorly. On each account I took the decision to simply yank the undocumented or poorly documented "elegant" stuff out of the system replacing it with something which average developers were able to explain and understand. The decision might have resulted in worse design, but who really knows. A customer who is going to maintain a solution deserves to be able to understand the architecture and design without having to hire A-grade coders and getting them to document the stuff (fat chance of getting A-grade coders who also document except for "Gamma or Beck"!). This culture of *just read the code* isn't flying today and brilliant people who can't express themselves are finding themselves having a harder time getting new project contracts.
Gartner, IDC and WhatHaveYa.. What is the actual value of these companies?
For the last 10 years I've occasionally watched some figures they spit out or some predictions they've made. For the first category, these figures are based on common knowledge (ordinary public statistics) to anyone in the industry, with a twist. The take a very vague number and from that try making it exact. For example, if statistics say there were about 1 million HDTVs sold last year, 0.5 the year before and 0.25 sold the year before that, Gartner will report 1,023,791 HDTVs sold last year and a "prediction" of 2 million sold next year.
As for the latter, I'd wager 90% of their their stats for new tech is complete bogus, missing the mark either totally or by light years. So what the heck is the force keeping them in business? What gullible people buys their trash?
As for their "report" on searches, again.. How the heck do they measure that? How did they come up with 36.9%? An what's the importance of that information? None at all as far as I can tell, other than for advertisers (who should get fired anyway if they listen to junk reports like this).
Slashdot Mirror
I agree. Likewise should the US government forbid acquisition of any tools, machinery, parts, components or other tangible or intangible entities which might conceivably hide some kind of intelligence intercept mechanism if said items are not manufactured in full within the borders of the United States of America. It's simply not worth the risk.
Come on, seriously.
I agree with you to an extent. However in my experience (having worked as a consultant for 10 years with mainly larger companies and a few smaller in various countries) I'd say the the platform adoption varies drastically between countries in general as well as for different markets (inside as well as outside a countries border to make things more complicated).
The highest Microsoft penetration per geographic area I've seen has been the Nordic region of northern Europe. The lowest in south america. For each country though, the pattern has been that larger companies are mostly using non-Microsoft platforms in the server park (Even in the Nordics except for their government organizations).
Now, in my experience there are two primary differences between larger and smaller companies when it comes to vendor selection and acquisition of vendor products.
First off, big companies usually have rigorous standardization in place and IT policies which often dictate what exact product should be used for what type of service in the enterprise. This means that a new vendor has a real uphill battle to sell products to an enterprise if that product is solving the same or similar problems as an already deployed one (from another vendor). The reward on the other hand if the vendor manages to convince the enterprise to switch is enormous since the policy will likely be changed to mandate the new product for future acquisitions and deployments instead of the old one.
Secondly, enterprise customers are very picky and have a stack of pages listing functional, non-functional requirements page up and down, many of which will not be matched by any vendor's product. This requires the vendor to adapt the product (by itself, together with partners or less often together with the customer's IT-staff) in order for it to pass the check-list criteria. Now in order to get the advantages of the previous paragraph, the vendor must be able to pass this checklist and given that very few vendor packages out-of-the-box(es) I've seen have been able to meet all checkpoints of any major client's list, there is bound to be need for custom software adaptations. AFAIK, Microsoft's business model is not to adapt their products to specific client needs and neither to build them in a way which allows them to be easilly adapted, integrated with non-MS products (of which most will be in a given enterprise). Companies like IBM on the other hand knows this game very well and make sure they can integrate with pretty much any other vendor's product they'll encounter in the client's server park. IBM which is also mostly a service (consulting) company has the manpower and skills needed to do whatever adaptations are needed to their offerings.
As I see it, Microsoft has a real challenge before them. In my view, they must also become a service company in order to compete with the level of service enterprises have learnt to expect during the past decades. However I'm not really sure this is aligned with the future vision of the company as Mr. Gates sees it, which might prove to be a dilemma as well as a non-insignificant hurdle to overcome.
What I was referring to was the implementation, not the ideas governing the implementation.
Take two different implementations of the same software idea and you'll surely see two different implementations. This even stretches as far as to implementations of the same idea at different points in time by the same developer.
Don't mend what is not broken rarely applies to software developers (unfortunately). Component reuse was a buzz word a few years ago which unfortunately didn't have much of an impact for inhouse development in cross project form. Developers like to experiment with new stuff and applying new algorithms or patterns to a known problem seems to keep them happy (avoids them being turned into factory workers i guess) and gives them a way to further educate themselves. However, this often impacts the deliverables in a (from a product manager's perspective) negative way since it introduces more unknown factors and deviations with each re-implementaion of a problem solution, than would seem necessary from the inception outlook.
Engineering in other segments (like the tangible sectors) of the market is simply more strict than in the software market, the latter where people build castles from thin air (pure thought stuff). In software the cost of trying out new approaches to problems is significantly lower than in the tangible markets which gives bored programmers a whole new set of freedoms when it comes to experimentation.
On a macro level though, the same governing rules apply as in the tangible segments. Stick with a proven formula since it minimizes risk. Yeah, it's boring and yields unimaginative products but that's what you get when the Harvard grads are sitting on the money bag.
The point I was originally trying to make was that when you pre-order stuff, you do it when you pretty much know what the product will be like. Final products can be predicted pretty well in the utility market but not in the entertainment segment. Utilities are built, marketed and sold upon proven formulas and solid history while entertainment are (or should strive for) about providing new experiences which may often require unproven formulas (higher risk and subjective in appreciation which if appreciated by many can bring in the jackpot). Since entertainment is based on subjective experiences it lends itself poorly at pre-ordering strategies from a consumer's point of view (the unpredicatable nature of the result that is).
Because it's better than simply reading the title and a summary at Amazon.com ?
Really, what's your point? People pre-order cars since most cars are just new revisions / bugfixes to older models with very little changing over each revision (such as the yearly increments of the BMW E46 model for example). I don't think the car business and their merchandise can be compared to the software industry and theirs. Programmers prefer to re-invent the wheel far more often than any other engineering profession.
The gaming industry is a segment of the entertainment industry and as such the same rules governing other practical / utility-industries do not apply. If I pre-order a non-software utility gadget which builds on an existing model (which is often the case), I know pretty much what I'll get. With entertainment this simply isn't true. Thus and apples and oranges comparison.
Today most laptops are just dragged between electric outlets since they wont go for more than a couple of hours without recharging
Agreed. Laptops are mostly used as portable laptops from my experience as well. This however is not affecting productivity as far as I've seen. At work we typically roam from room to room with our AC-adapters for meetings, presentations or getting some work done in the labs. Coupled with this we also roam to different client locations which all have power outlets for our adapters. Point is that the pattern is people moving from location A to location B and spending the majority of time not moving around. When stationary, plugging in the AC-adapter is not a big deal.
However, compare this to a full blown stationary setup with a 15Kg case and a 7Kg monitor, not to mention all the associated gadgets such as keyboards, mice and cables, the true benefit of a laptop for a person who isn't completely stationary at one site becomes clear. A all-in-one computer to bring with you to each new location.
Battery really isn't that big of a problem from what i've gathered from clients in most industries. What people in my experience clamor for is the same stuff as desktop users want; More speed and, better displays and more storage.
Sure the ongoing 10% yearly increase in battery life is nice, but other attributes take precedence.
PS. Imagine the new logistic problem with fuel cells for companies with multiple sites. "6 hours" (9 hours and subtracting the 1/3 optimistic marketing figure addition) on each kit, meaning I'd use about 2 a day. At our office alone, we'd have to stock up with 8000 each week. The company as a whole would burn through about 1.6 million a day or about 350 million kits/refills a year... For one company alone.
A thought just crossed my mind.
Since admins running Unix-like systems mostly operate as non-root users, wouldn't it then be possible for a malware to lurk in the background of the non-privileged sandbox until you sudo/su and then for it to use the newly gained privileges to wreak havoc/gather intel and hide itself? In a non-root sandbox the malware process would likely show up in the process list, but who can honestly say that they check the process list each time before they become root? Also, a malware naming itself like a common process (or the same as a process which already occupies the list) like bash for example, would make a casual glance likely to disregard that listed entry.
Disclaimer: I don't pretend to know the intrisinct details regarding privilege escalation in *nix, so this thought might well be nothing but nonsense.
First off, I think you totally missed the point regarding what Trusted Computing really is all about. Giving ultimate authority over stuff you buy to the vendor instead of you (the user/consumer/prosumer/sucker buying the stuff).
Secondly, you talked about something else, namely information loss and that you claimed you could prevent it. Here your arguments are vague and rather lacking i.m.h.o. The only secure system is a system without external interfaces. Such a system is on the other hand not very practical, so essentially all system have them. Without knowing what you use your systems for or how, I will venture a guess that some of them at least are networked. Further, if you with these systems provide any kind of information to the outside world (e.g. web servers) then there are countless schemes to get information out of your machines without you having the simplest notion the occurrence.
Example, say you get hit by the "latest" Apache / IIS / Bind or WhateverApp virus, which happens to modify said app to do malicious information gathering. After having gathered the information it can then use your compromised application as a relay to allow the virus author to obtain the information (e.g. through some protocol like for instance inside a specific web page in base64 encoding, specially crafted ping replies or whatever). Point is, if someone is able to compromise your applications on a networked machine, then there are countless ways of getting that information out without you having the slighets clue.
Note on boasting; Hubris is what brought Icarus to his death after all.
Well, in Europe right now, there is a big conversion going on towards GNU/Linux.
I've personally been involved with the switch of three government organizations in two countries on the server side and one on the desktop. From colleagues in other countries (esp. Germany) they are succeeding to migrate to a better platform even faster.
As more people become accustomed to alternative platforms and applications, they tend to use them at home as well. Nothing substitutes learning like the massive hands on experience at work or in schools.
Would MS be forced to exit the EU, it would probably be under a grace period suitable for an accelerated program of rolling out alternative platforms and tools. Not over night, so I would not be worried.
The problem is transparency.
Would you stake your business or for that matter, you life (as is the case in some regions of the world) on this assumption? Since there is no transparency in Microsoft products, you simply have to take their word for it.
I thought the golden rule of security was that any viable security mechanism should tolerate public scrutiny. Knowing how the software works should not work against the devised scheme itself.
If this isn't vapor ware, why in G*d's name do they alert the competition of this fact? I thought the normal practice was to slam your competition suddenly and without warning, which in my book doesn't include touting future plans before you're forced to do so.
Unfortunately, my modpoints are all out at present, or I would have rated your post insightful.
In my view, the US government is only listening to one voice, the voice of military offensive power.
In that light, since it's proven mostly futile to reason with the US regarding important (as well as less important) subjects, perhaps the rest of the world should aim at restoring some balance of power. This is really the last thing I would like to see happen, but the US really doesn't seem to be giving the rest of the world much of a choice. Some ramp up in offensive power in the EU as well as China should hopefully counteract some of the imperialistic agression of today's United States of America and help draw some much needed lines.
"Enough! Here, but no further!"
Yes, treacherous computing will keep the digital roads safe and locked down. Pay or we exclude you from using your own computer.
All information will be "pay-per-view". Step one is to ensure nothing ever again sips out into the public domain. Step two is to protect the concept of every idea and make people pay per use for each. Instead of "thoughts are free" the saying will be "thoughts might be available for a pre-negotiated price". Without enough money, your child won't be implanted with the genes required for a concious mind. You'll get a vegetable.
We're heading straight towards this future. I really do hope people remember why the right to keep and bear arms was so important... Someone better come up with a good alternative idea (and fast!) to change the current horrid path.
"If you don't like it, don't buy it". This advice is becoming rather repetitive. It's also not an efficient way of "signaling" to a company that the XYZ they do sucks and they should stop.
If someone regularly (once a month?) spends around $100K or $1M on stuff from a given company and suddenly stops, then that might raise an eyebrow and perhaps trigger the company to question why the person stopped the regular spending of money. Anything less will *not* react much of attention.
On the other hand, a person whose opinion is voiced (blog, newspaper etc.) will generate a *much* larger impact. Especially since most larger companies have entire departments scouring the web and other media for tidbits about them. Also, by voicing their opinion, the company gets to know *what* they did wrong. Just stopping the money spending tells them little to nothing about what's wrong.
So, don't kid your self that a single person "not buying an i-pod" for $200 will result in any specific change at companies like Apple.
Perhaps this "fury" over voting on the "other guy" could be addressed by having more candidates?
Now, I realize it's in the politics 101 book to always envision "an enemy", since it's a tactic for making people who would otherwise not agree come together. However, as I said, it's mostly a tactic by politicians *for* politicians and eliminating or weakening this tool would be a *Good Thing* in my book.
So, again, why not introduce more candidates? Also, why not count each single vote by each single voter cast in the election? Today's state lumped voting mechanism isn't really working in a democratic election when the candidates are from the people (any given citizen deciding to run) as a real democracy would allow.
Finally, the argument that "people are too stupid/lazy to keep track of more than two things at a time" doesn't fly with me. Even if you had 1000 candidates, there would be people who would not vote/care, same as it is in today's system trading whitches for devils. It doesn't mean a system with 1000 candidates is any less democratically efficient than one with 2 candidates. More candidates simply means higher probability of finding a more suitable candidate in whom you would entrust your next 4 years to. This would likely result in greater number of active voters since a suitable match would be more likely. I think one of the current problems is that neither of the current parties appeal to a lot of people and so people simply choose to not "get involved". This is no worse than would the situation be had these poeople arbitrarily picked a candidate at random for voting on, right?
More choices equals more freedom and allows for greater amount of expression.
If done responsibly by the ISP.
I'm fortunate to have a 10MBit symmetrical connection from my ISP, soon to be beefed to 100Mbit. I also know that the ISP is shaping some P2P protocol traffic of which bittorrent is one. However this ISP realizes who it's customers are and that someone ordering 10MBit internet access is probably going to use it to some extent since they didn't go with another provider offering less capacity. So, the ISP has a pretty good infrastructure in place with good peer agreements with other ISPs. Now, even though they have a fairly well developed infrastructure they will run into extreme peaks and must be able to manage dataflows so that the main internet services are available to their customers (telnet, SMTP, VOIP, HTTP/S, SSH etc.). So they use traffic shaping to simply assign lower priority to P2P traffic.
This behavior is totally OK with me and I wouldn't have it any other way. Certain services are more time critical than others and I wouldn't like them to be affected by the huge P2P clogs in the network. On the other hand, since they just re-prioritize the packets I know that my P2P transfer rate will be good enough (the remaining capacity of the network when the essential services are cared for).
This solution works great when you have an ISP which continually beefs up their backbone, but would pose a problem if you have a cheap *ss ISP which models their backbone capacity only for the (by them considered) essential services.
So, in short, if you have an ISP which do cater for your specific customer type, then traffic shaping can be a good thing for you as well as your fellow ISP customers. Traffic shaping isn't all evil, that's all.
When I write a program for myself or in a team with just a few developers who know how every aspect of the solution works, it's nice to change stuff at a whim. Since the team is small, everyone monitors each cvs-update for changes and talks to each other frequently to keeps abreast of what has changed. However, for projects spanning more than just a handful people, specs quickly becomes essential as a communication tool.
An API is just that, a communication instrument (or protocol) which is intended to describe the exposed interfaces, their purpose, what they do when used in various ways etc. Essential information for using a component. This is information any user of the component would need to know anyway in order to use it properly.
Now, when you mentioned refactoring being sacrificed if interfaces (APIs) are published (which is what I interpreted your text to say), I completely failed to see the underlying reasoning.
(reference). What this seems to say is that the whole point of refactoring is to keep the code from decaying while not changing the interfaces!
If you require new features / behaviors, why not simply create new function signatures and deprecate old ones? What's so difficult or bad with that approach to changes?
Somehow I don't think that's really what the parent poster meant. ATI could very possibly provide passable support if they wanted to, but they don't. ATI is notorious for their total lack of interest in Linux. I agree with you, that any sane company aiming at establishing a foot hold in an emerging platform market would jump at the opportunity to make their own products work great on said platform. However ATI is not such a company. Sad as it might seem, ATI still doesn't see enough value for providing adequate product support on Linux. They're suffering from the hen and chicken syndrome, hoping that the effort spent on RedHat and SuSE (since that's what Linux basically is to them) by their competitor(s) might some day make it grow to a viable market segment for them.
Why would anyone be excited at being an early adopter? Isn't it common sense to let a product reach a certain level of maturity instead of playing the part of a ginnie pig / extended QA arm of a company, having to pay for the displeasure of finding and experiencing kinks in the offering?
Observe and adopt when the time is right I say... When the flawed fans have been replaced, the heating problem resolved, the availability of enough accessories to add proper value to the offering etc. You know, the typical scenario for new products.
People don't care about weird names.
They do.
Ever looked at the credits after a movie?
I'd wager 1/3 to 1/2 of the actors in the list have changed their name. Either because their previous names were too convoluted for the average American to pronounce (e.g. they might have ancestral ties to other countries) or they had too common names, meaning they needed to change them to stick out just enough.
This example simply shows that names are very important to us when it comes to instances of types.
When it comes to the specific types (e.g. types of technology) themselves however, we tend not to care that much what names we use to classify the different things with. Since many of these things stem from science (discovery) we might lack prior art and have settled with the fact that "new stuff will yields new names". Our perception of the new stuff is formed by what it can do for us (after having seen or read about it) and so "strange" names like Turbo, Laser, Gasoline, Homo sapien etc. get established and recognized by many as types.
What would help tremendously though when naming new stuff (types) would be to look at the traits of the concept and pick a name communicating the intent of the type. This would help people quickly pick up what the thing is about and speed up awareness immensely.
E.g. Say you invented a strap which you could tie around a dog's neck and also attach a leach to. Now instead of giving that thing a name like Smorgasbord, wouldn't a name like Dog Collar be easier to remember? Wouldn't it also convey better the intent of the device to people not yet familiar with this new thing? The concept collar is already familiar for most people and to distinguish this new type variant from the one humans use, adding the type name of the intended wearer will avoid any confusion in the morning.
So, is Joomla a good name? Perhaps not. Will it be catastrophic? perhaps not. Could a better name have resulted in wider awareness in shorter time? Personally I think so.
If you manager produces noticeable value in the area for which s/he was hired (technical or non-technical) for you, your team and the rest of the company, then s/he is probably in the right spot.
Managing and being at the fore front of technology is VERY taxing and tends to result in burn outs of most people. I've seen people being able to cope with such jobs at a maximum of 18 months at a time before requiring downtime (e.g. focusing on either for a while) or they'd be hitting the wall (too many have gone down the latter road for my liking).
Personally, I've been in a position where I for the last 12 months have had to negotiate contracts, make planning, acquire resources, deliver architecture and manage the execution and delivery of systems for a government contract. If it were not for the fact that I'd be handing this over to a replacement in 1,5 months I'd burn myself out.
When I eventually do take a position in a line organization I'll definitely choose to *either* focus on sales, resource management, project execution or architecture. Not all of them at once since it's unrealistic to expect an "IT Manager" to keep "consultant" like pace as part of a long term career (5+ years?). No person I've ever met have ever managed such a feat and expecting such from your manager is not realistic i.m.h.o.
And yes, if when the time comes to choose a path and the path happens to be non-technical, I will become less skilled in the areas not being the core of my new position. E.g. I would not see myself being able to teach younger specialists (as an example) the ins and outs of various techs. I'll expect any person I hire or manage to be better than I in their respective field and will rely on their advice and expertise for making the proper decisions.
Please reflect over your manager's situation and the work s/he does. In most likelihood the person is busy making sure you and your team can perform to your best while at the same time "shielding you " from upper management and the issues coming from that direction. I'd also wager SLA , customer commitments and allocated budgets weigh heavy on your manager to name but a few examples.
At a talk RMS held in Germany a year ago or so, he mentioned that a *friend* of his had gone through the Linux source in an attempt to find possible patent infringements. RMS mentioned his friend found thousands, so regardless of Perens being a d*ck or not, there might be some substance to the claim.
Typo...
"Therefore I hope Google will make it possible for users to include iTunes hits".
Should have read
"Therefore I hope Google will make it OPTIONAL for users to include iTunes hits"
Not everyone is hyped about the iTunes service. Therefore I hope Google will make it possible for users to include iTunes hits. Otherwise this deal results in instant degraded search results for a lot of users (those not interested in itunes).
Personally I despise the service for one major reason. The quality of the purchased sound SUCKS! To be fair and giving Apple *some* credit, the packaging (the application with it's simple user interface and it's preview functionality) is nice. However the wares being sold is of lousy quality and the primary point of iTunes was selling sound (at least what I though).
I also wonder if the EU has a case against Apple for errecting trade barriers. Not all songs are made available to all member countries of the EU which is another thing deeply disturbing to me (some songs for ex. available in the UK store are not available in the German).
This seems like an excellent book and subject.
The industry is in dire need for more people to write for the *need* of other people.
Two current examples.
My company employ one of the big development methods, aging back some 10-15 years. It contains templates for various (document) deliverables the company pitch as an advantage to customers. These documents (ranging from 80-500 depending on the project) are something each project spits out *just because* they have to (lousy project management). The decision of what document deliverables to produce is made something like 1-2 years ahead, which in reality means only 5-10% of them would be useful at the date of shipping.
Instead of having all these tens of thousands of useless crap pages being manufactured, junk which no-one will ever read, it would be much more appropriate if the right people could write about the right thing in a way suitable for the intended audience. Unfortunately this would mean a break from the *method* at various point which seems unlikely at present given the particular company culture. In short, Agility needs to enter the "document manufacturing" cycles as well.
The second example has to do with with doers vs. explainers (is that a word?).
I've run some projects in which we had some brilliant designers and coders, creating the most flexible and elegant designs and implementation. Unfortunately these people either didn't document them or documented them poorly. On each account I took the decision to simply yank the undocumented or poorly documented "elegant" stuff out of the system replacing it with something which average developers were able to explain and understand. The decision might have resulted in worse design, but who really knows. A customer who is going to maintain a solution deserves to be able to understand the architecture and design without having to hire A-grade coders and getting them to document the stuff (fat chance of getting A-grade coders who also document except for "Gamma or Beck"!).
This culture of *just read the code* isn't flying today and brilliant people who can't express themselves are finding themselves having a harder time getting new project contracts.
Gartner, IDC and WhatHaveYa..
What is the actual value of these companies?
For the last 10 years I've occasionally watched some figures they spit out or some predictions they've made. For the first category, these figures are based on common knowledge (ordinary public statistics) to anyone in the industry, with a twist. The take a very vague number and from that try making it exact. For example, if statistics say there were about 1 million HDTVs sold last year, 0.5 the year before and 0.25 sold the year before that, Gartner will report 1,023,791 HDTVs sold last year and a "prediction" of 2 million sold next year.
As for the latter, I'd wager 90% of their their stats for new tech is complete bogus, missing the mark either totally or by light years. So what the heck is the force keeping them in business? What gullible people buys their trash?
As for their "report" on searches, again.. How the heck do they measure that? How did they come up with 36.9%?
An what's the importance of that information? None at all as far as I can tell, other than for advertisers (who should get fired anyway if they listen to junk reports like this).