What the T-series really needs is a boost to per-thread performance since it will otherwise remain a specialty processor only suitable for certain workloads.
The T2 core has more than enough parallelism for most apps out there. What isn't appreciated though is that it pushes the server *implementation details* all the way up to the app-developers, which causes them grief when they need to target different hardware or when they utilize "junior" developers. It also causes a lot longer performance tuning phases than on our previous platforms (SPARC and Intel).
This situation is fine for A-list developers but causes major grief for the multitude of companies whose developers don't have experience in massively parallelized systems. Companies in that situation unfortunately are most out there.
Our local SUN.. I mean Oracle drones always point out that our servers are able to handle so much in parallel, but that means squat if we can't meet our SLAs.
If I have a response time SLA of 1 second, then it does me little good if I can service 10 times the number of requests of a competitor's server if each request takes 3 seconds and the competitor's hardware actually allows me to reach the SLA!
Also, if there is any kind of locking going on, the server will more or less halt and whopty doo there goes the parallelism.
We won't be buying another T2 since event the PHBs can read the productivity charts and risk reports handed to them by external consultants. The cost of performance tuning of apps has climbed a lot and cost us a small fortune and continue to do so. Hardest to cope with in this space are legacy and third party apps where hardly anyone dare update the decayed code (or receive funding to do so) or in the latter case can.
However, we're seriously contemplating buying a bunch of Nehalem-EX servers and would perhaps have bought the Power7 if we were an IBM-shop since both those companies "Get It", get what customers as us need in contrast to Oracle.
Advice to Oracle: Add a bunch of cache and allow for higher clock speed in the T3 to really start competing with Intel and IBM. I don't care if you add a thousand more threads if each thread still incur a latency three to four times longer than your competition.
Finally, yes I realize I come across as a sour grape but the amount of time and cost *wasted* as a result of our PHB buying these servers (based on spec-mark figures) without contemplating the intended workload has really put a dent in my department's work atmosphere.
Well C(++) doesn't force you to use new lines at all for most purposes. In the PP's example the exception would be the pre-processor directives. So, the PP's solution would fit in just 4 lines of code, way fewer than the 60 required for the Python example, but what does that prove w.r.t. either language implementation of the algorithm?
Lines of code is a real stupid measurement for complexity, productivity, maintainability and / or elegance and I really wish people would stop using it like that.
You left out one important fact, that enforceability of patents in the EU is up to each member nation to decide on. This little fact you forgot to mention means that software patents are null and void for the majority of the EU states.
The sole exceptions seem to be the 52rd state of the US, commonly referred to as the UK and to a lesser degree Germany.
Still, both German and UK courts are in a great flux regarding the entire software patent issue so the overall state of patents in the EU is that they are more or less a complete waste of money and time for companies to acquire.
For that reason the PP makes a good point. If Canonical made a European version which ignored the legal minefield currently plaguing the US software market and offered it up to EU citizen with a disclaimer that anyone else may download it at their own peril, then hopefully that would start a debate in the ROW (rest of the world) regarding the damage / benefit balance of software and process patents.
Unfortunately, politicians have neither the insight nor the time to ponder about hypothetical / abstract situations which means the most efficient way to get through to those people is to *show* them something concrete. When the ROW-politicians see their citizen illegally using a distro variant more capable than what is permitted in the home country, a distro which is legal in other countries btw., they will start pondering why the state of affairs is as it is and the damage the current situation inflicts on their society and economy.
It will not be an overnight change but it would most likely be another spark in the kindle to get a sober discussion going in nations which have completely capitulated mentally on the issue.
You could aim at becoming an operational architect, a person who decides what boxes to get, their configuration and amount, network topologies etc. Several colleagues with that focus haven't touched a line of code in years.
Assignment scopes range from small (e.g. supporting an internal application with infrastructure) to large (enterprise wide architecture spanning multiple data centers). The job includes a lot of requirement mgmt. and specifications. Some project management work is also implied (e.g. TCO calcs & comparisons, rollout planning and staffing..)
The pay is very good, but you may have to seek out larger firms where the complexity to be managed justifies having architects with specialization in the functional *or* operational domain. At smaller firms (read less complex environments) the same architect typically performs both parts of the overall design.
However, if you don't already have a lot of experience with boxes, networks, security and lifecycle management of applications then a good start would be to get that first. As someone else mentioned, networks admins, DBAs, server installation etc. are accessible areas for junior people.
An entire additional nation now needs to be able to post information to Slashdot without risk of reprisals from the government or corporate interests. As such I would like to emplore the Slashdot administrators to enable SSL as an alternative to un-encrypted HTTP traffic for reading and posting to this site.
I am fully aware that SSL will increase the resource use of the site, but if you make it a feature that must be enabled in a user's profile, it wouldn't be a default and thus the performance impact should be manageable. As we all know, anything requiring "opt in" will mean only a fraction of the total population will use it.
If you can spare the CPU-cycles, a good service would be something akin to Google's, where you enable SSL for certain (surveiled) IP-ranges where as Google uses it to "i18n" their pages.
Revised and re-voted within just a couple of hours should be pointed out.
The instigators of this constitutional violation had done their homework and were armed with the appropriate fallbacks (contingency tactics) including revised paragraphs and buttons to press to get the political party whips flying so as to eliminate any self-thinking representative from diverging from the party flock. That is, get the slav... ahm. dron... shee.. I mean elected congressional representatives in line for a second vote on a revised legal proposal. Focus was clearly to just get it through without having it properly reviewed and disclosed to the public, thus the prepared amendments.
Everyone but FRA and the party leaders of the ruling council was against this motion including the opposition, the secret service (SÃPA), the legal establishment, the journalists and not to mention the people of Sweden. Still the mission was deftly executed, I must admit, forced through in record time not even breaking a stride when it was rejected in the morning. Well prepared and executed and kudos goes to the evil masterminds of the plan for a "job well executed".
This is a sad day but one which hopefully has opened up a number of important questions regarding the extent to which Sweden claim to be a democracy. Even your average dumbed down person on the street has started to realize some of the implications of STASI-like surveillance. Other questions which have arisen as part of this fiasco is what value having named candidates on the ballot really has, when at the end of the day those people will still be threatened into following the party line regardless of the opinion of those who hand-picked them.
So what now? If you have a need of communicating with people or organizations in Sweden, require of your Swedish counter part to use encryption for ALL communication. For any company having information infrastructure in Sweden, please consider moving it at least as far as to Finland (an adjacent neighbor with as good an Internet infrastructure as Sweeden has) where this kind of mass surveillance is prohibited. The ISP and phone provider TeliaSonera is already in the works of doing this and that should provide a cue for the rest regarding the gravity of this decision.
The government has all but admitted it intends to use this new tool for corporate espionage as well, to get at inside-information regarding interest changes etc. Not a good situation for corporate interests, unless you're the highest bidder and thus gain more inside information than what you loose to you competitors using the same machine.
As to the comments regarding external threats, well that's just pertaining to the analysis phase. The FRA will gather EVERYTHING from the channels / protocols they deem of interest and will LATER analyze that data during the activity defined as signal surveillance for whatever they're missioned to search for in THAT context.
What hasn't been widely discussed is the fact that FRA has a number of interests which can issue orders to them outside "signal surveilance", including the police, the ruling council of the government, other branches of the military as well as certain private interests. These work orders are executed based on the SAME data being intercepted as part of the "surveilance" directive (I.e. all data). In the future more interests will likely be able to get at the data stored by the FRA and given the amount of people having access to this data, when leaks start becoming more prevelant the floodgates simply are no more. (As an example, the road charging system hadn't even been completed before the police came knocking on the door wanting to make use of that info, eventhough not permitted by current law. Think this will be any different?).
PS. The FRA has ILLEGALLY performed surveillance for many years in areas prohibited by law but that's been constrained to signals in the Ether / air. This new legislation will REQUIRE all ISPs to clone their packets and route the cloned traffic to FRA for storage and later analysis. That's a HUGE scope change and the impact is hard to fathom!
Any extension of copyright by the US was most certainly *not* to maintain parity with the EU but to keep control by Mickey Mouse (read general greed). There is no general duration of copyrighted works in the US because the US copyright system is a mess
Did you read the article? The system is claimed to be anonymous and I tend to believe that since your personal data (who you are) isn't broadcasted in the operational signal but reside in a database owned by your specific ISP.
The system seems to be tracking unique phones by signal characteristics, thus allowing the phone's path to be plotted but the data delivered is still just "some phone took this route through the monitored area".
And to limit the "DRM" impact, making the GLONASS, Gallileo and GPS systems transmit both CDMA and FDMA signals, receivers supporting the two formats would essentially eliminate the "DRM" fear.
Sure, each individual player (US, EU or Russia/India) could selectively disable their own system's coverage over a region if they wanted, but the receivers would just pick satellites from the remaining systems.
So it seems rather redundant and costly for a system to incorporate regional off-switches when the power to select satellites is still on the receiver's side.
Asynchronous Javascript and XML (or in practice XmlHttpRequests using any kind of text based wire format) are a nice addition and if used as pure decorators on top of sites driven by server side markup generation, then it is a good thing. The issue I and others have is that what "Ajax" seems to also imply when applied by sites in practice is a heavy reliance on a lot of client side DOM manipulations.
A site designed around the notion that as long as "Firefox and IE" can morph the bootstrap HTML page through an infinite number of morphs to something completely different, then the site is good. Those kinds of sites are neigh impossible to use via Wget, perl, lynx or any client not having:
1) A Javascript engine 2) A DOM engine 3) a special variant of A and B combined in such a way that they replicate the same quirks (attributes and behavior) inherent in IE and Mozilla.
So to sum it up, I don't think anyone has anything negative to say about requesting data fragments as an alternative to doing full posts/gets to the server. It's when people are being forced to one of a select few specific applications in order to use the web that irritation starts surfacing.
CPUs, memory and storage are today getting to a price/performance point where it starts making sense for manufacturers to contemplate building consumer devices on top of a scaled down but familiar platform with few or no license fees associated. Chinese manufacturers are I think leading the way in this area atm. and we'll soon see the fruit of that interest.
This is a good thing. Consumers are many and varied and most of them are non-techies. To sell to non-techies you have to really nail the (user interface) experience and lessons learned during the next 5 years will eventually trickle back the desktop domain.
So "Linux on the desktop" will i.m.o. not be something that will happen until Linux is in most of our tiny devices (iphone/ipod clones, nokia phones, portable media centers, wearable GPS devices / personal network hubs and whatever other gadgets of today and tomorrow.. ). So my guess, 5-6 years before we start seeing Linux widening noticeably on the desktop, but at that point the current obstacle holding Linux on the desktop back will have vanished and then it will be the final time we see an article on./ titled "Is this the year Linux will conquer the desktop?".
How can we as developers know when we are working around an intended limitation and when we are simply adding "value add" to a product?
If I create a multi-segment downloader extension for Internet Explorer, am I then working around an intended limitation in IE and will I get sued or is it perceived as something which enhances IE and thus avoid a lawsuit from Microsoft?
With this action they've really shown that developing for their platform (they own all the base components) is a very risky business for any developer except their own.
Unfortunately I'm not a lawyer and as such may have misunderstood the three laws in play here, but according to this interview segment with Eben Moglen, it sounds that his counter argument will be that through the patent deal with Novell, Microsoft knowing full well of the imminent introduction of GPL V3 knowingly extends their patent "protection" as per the provision of GPL V3. Now I agree, his argument doesn't clearly address the main concern of your analysis but to me he seems to rely on one law (copyright law) being able to affect another (patent law) through a third (contract law) and that to me seems to be a very risky position to take if not just very complex for non-lawyers to wrap their minds around.
It would be great if you could send your analysis to FSF in order for them to provide their reasoning around how they plan to tackle the question. If you do, please either add any reply you may receive to your/. journal or pm it to me since I'm very interested in their response.
Can a copyright license control patent provisions? Can aspects of an agreement made (contract law) between a party and another be automatically extended beyond the agreement by actions performed and circumstances surrounding one of the parties for actions and circumstances not covered by the agreement itself? Muddy water this, to be sure.
Well, these european countries (norway, sweden..) don't have laws which prohibit cracking or circumventing DRM so that's not an issue. However, these "bans on DRM" encourages hardware manufacturers and distributors of DRM-shielded data for these devices to provide this restriction-less behavior "out of the box" for average Jane to enjoy.
DRM circumvention schemes are still rather technical and requires more work than simply plugging in your ipod and moving the music to your hard drive for enjoyment on your connected Hi-Fi equipment.
Well, judging by other countries it's likely they will retain the levy and still make it illegal to copy stuff. Taking Sweden as an example, there they tax all storage media (not just "removable" media) with this levy. Actually it's not really a tax (taxing illegal activities is illegal itself), but a state protected fee which a private organization is allowed to collect and without insight into how, distribute parts of the sum to a secret list of copyright holders.
So I guess you have to look forward to being coerced by your government into paying levies (which should exempt you by covering any IP loss right?) and then be put behind bars. It's a hypocrisy, but hey at least it isn't the first in the law book.
It's seldom a good summary is posted on Slashdot. Typically the poster provides some nonsense or poses some arbitrary unrelated question at the same time. I hope more posters take after kdawson in actually reading the article, digesting its contents and add some value to the slashdot readers by exposing the gist of the article.
We've got too much crap-publishing going on and anything and anyone who helps reducing the need for readers to weed through crap is a good thing.
Hmm, I've been reflecting on the responsibilities of manages, PMs and architects for the last few weeks and my conclusion after about 10 year in the business is this (after an initial outline):
Architect An architect must know the business well, so that the the problems in the domain contexts can be identified and resolved. The person also need to know a wide set of technologies and principles so that a viable technical solution to the business problems can be devised and managed. The most challenging part of an architects job is balancing the various forces/needs/desires/priorities, such as money, features, quality, time as well as factoring in the available skills pool containing the resources (people) who will realize the solution as well as those who will maintain it. Already at the high level PM-like activities are performed such as creating time estimates for each component and dependencies between the component deliverables are created. This is then mapped to a project plan with activities parallelized as needed to fit the current realization constraints (e.g. time vs. cost) which is handed over to a project manager to "execute" (follow up). A person having this role the flack is the solution doesn't meet the requirements.
Project Manager A person who more or less sticks to and revises a Microsoft Project plan which is created to a large extent either by or together with the architect. The PM also manages the internal bureaucracy when it comes to hiring / staffing teams and ensuring the budget is factored into the realization of the project (e.g. if unforeseen events such as delays or mis-calculations strikes against the budget, a remedy activity must be initiated). The PM is more or less an administrator which is hopefully following some method for project delivery. Saying that a PM is an administrator is not a degenerative statement because every project *needs* at least one administrator to off-load and shield the project team from external turbulence and activities not relating to the progress of the realization of the solution. This is essentially ensuring that the cogs are greased so that the machine works smoothly. A person having this role gets a whopping if a project doesn't meet the dead-line / runs over budget.
Line Manager (which most people tend to refer to when they say "manager") A person responsible for hiring (and firing) people to the company. Administrates the yearly assessments of employees and helps set individual development plans for the people s/he manages. Is also responsible for distributing the yearly bucket of money for salary increases to the team. Is an echo-chamber for the manager above him/her. First-line managers are the most common and less desirable positions in a company because at that position one has to echo unpopular/stupid decisions "downward" to the people who are actually bringing in money to the company, while trying to manage the outrage from the "floor". Kind of like being stuck between a hammer and an anvil. A person having this job is hired to get flack from upper management or the managed people (all the way up to the exec. who gets hit by the board if the news are bad).
Now, a PM in the tech-industry who has a background as a tech-lead or architect would surely be more desirable than a PM who just graduated "PM-school" or a senior one who has never been part of the actual delivery of technical solutions. For a manager, it would perhaps not be as important as for the PM, but for the people being managed to be able to talk to the manager in situations where they end up under a bad PM for example, the manager would be more likely to understand the problems faced by those people if that mgr. understood the "language" and the set of typical project deliverables. E.g. A specialist who is in a crit-sit project who is over burdened with activities and who has an inexperienced PM might be able to resolve the situation by talking to their manager. In those discussions hands-on experience by the manager would help him/
Nearly as important is the fact that the support contract forces me to upgrade or else Microsoft won't give me support.
Windows 2000 was perfectly fine for me but seeing as it was retired last year I had to move to XP (couldn't afford the extended support contract. Cheaper just getting XP licenses).
how to store digital files so future generations can access them, from engineering plans to family photos
Well, why not transcode it onto a more sustainable format? Just create a device capable of understanding common printing formats (e.g. postscript) and let it store the data onto more proven media, such as, I dunno, stone or some form of metal...
We have printers for printing rendered data (text/images) onto papers. Why not have a stone or metal printer which uses a lazer with a bot more watts to etch the information into these durable materials? I concede though that searching these etched medias would be more of a problem, but there has been no system that I know of to this date which were capable of accessing vast amounts of information, packed into a small space other than digital/magnetic/photonic storage (comparatively transient in nature).
After the nuclear winter, the next civilization will simply have to "scan" the recorded knowledge of 21th century from rune stones into computers again.. We did it so why wouldn't they..
PS. I hear the moon is pretty rocky, so perhaps we can use it for something after all.
This means, for instance, that contractors can be allowed to attach to the enterprise network using their own laptops but only via the ACE VM
For all clients I've worked with, they either:
Allow me to use my laptop on their network and allow me to connect to my company's network through their corporate network or through a dedicated network they've set up on the side, using a VPN client.
Allow me to use my laptop on their network, but do not provide me with a network for reaching my company's network.
Do not allow me to use my laptop on the corporate network, but instead provide a "standard laptop", same as the rest of the workforce is using or a desktop to use.
In the first case, I'm always more productive and cheaper per unit of delivery than the other options, because I've got all my tools setup and on average all software (except perhaps one or two apps) I need to perform my work already licensed. I'm also able to harvest the information, knowledge and assets available through my company's network. Some clients realize the benefit of this and instead regulate which tools I may use and what I may do while connected to their network by making me sign an agreement (along with the standard NDA).
In the second and third scenario, clients still require me to sign the same "usage agreement" and NDA, so there really is no benefit for them from a liability standpoint. The risk is the same for all three cases and mitigated the same way (through agreements). On the downside, each unit I deliver costs more because I don't have my toolbox and don't have the software configured for optimal productivity. However I still deliver reasonable value per dollar spent because I am still able to leverage resources from my company by using my own laptop on the side (either through a client provided network or a 3G card).
There is really no benefit that I've witnessed from a client perspective in not allowing me to use my own machine and tools.
Now, the quoted statement of having me use my own machine coupled with the limitation of not having my tools available and also not being able to use the resources available through my company's network for the client's benefit seems like a worse situation for a client than any of the previous ones.
No matter from which angle I try comparing the three typical scenarios with using ACE for the suggested purpose, I fail to see the bottom-line benefits for a company relying on consultants / contractors.
I'm not sure I agree that it is a good patent and I do foresee problems which may well arise from this. The first 75% of the patent is a digest presentation of common knowledge coupled with all the domains for which they claim dibs (any networking domain or concept publicly available when the patent was written, from satellite communication to DSL and Cable) while the last 25% started to go in to the more specific suggestions of how they propose some sets of problems *may* be solved:
High availability (reducing the likelihood of packet loss between producer and consumer)
How discretionary provisioning of packets may be be realized (tagging data for consumers or to a set of consumers in a branch of their proposed network tree model, fanning out from producer to consumers, kind of like an LDAP directory tree).
The thing which bothers me is that the patent is a mix of a "for dummies" guide of a set of concepts which are obvious and have been taught at schools and at various tech-companies for over a decade, coupled with some very broad suggestions of concepts of obvious use. The "innovative" part (read. specifics of the proposed solution rather than "innovation", since it's arguable whether there is any innovation in those details as well) is both hard to isolate and account for perhaps only 3-4% of the patent. The result I fear is that this patent may be used to attack a broad range of people and companies delivering solutions making use of any one (or a set of) the 95+% common knowledge concepts which redundantly were incorporated in this patent application.
In my mind, this is a bad patent from the outlook that it looks more like a legal weapon enabling attacks on solutions deviced from public knowledge, while it's not providing enough focus or detail to be used as a blue print for creating a content distribution network (which seemed to be part of the aim). A good patent for patent lawyers and their income, but not so much for society and the economy at large.
I've seen worse, but this one isn't close to qualify being in the upper quadrant of beneficial patents.
Since Sony went "digital" (minidisc) I have mostly associated Sony with obscure audio formats (atrac), lossy transcoding and impossible to get my music off the portable player. With Blueray, the perception of a anti-consumer company is strengthened even more due to all the DRM crap for thri higher fidelty video format.
Piled on top of that your three examples and it's stupendously clear the company has as much allure as does the British kitchen.
The problem with applying the open source idea to an online encyclopedia is that this is _not_ a "Oh, there's a bug! I'll fix that so that it works as advertised". It's a "I don't _think_ that's the way it is, so I'll change it to _my_ reality." Where that reality is _not_ necessarily the real fact(s). And from what I've read, what facts can be very illusive on the wikipedia.
Well, I think it depends on the topic. If it's a scientific topic, then the truth can be proven (given the acceptance of a current framework of reasoning and deduction). If a topic on the other hand is more subjective in nature (like history tends to be or social topics) then I disagree that there are absolute facts. Recall the phrase "The victor writes the history". This alone implies there may well be a lot of factual errors in "official" or "published" literary works and in that context I'd not necessarily put more value to [enter-your-favorite-encyclopedia], than on a source where topics may get formed from the combination of multiple angles on the same subject. Examples of this would be topics regarding the "war on terror" or middle east politics which happen to be topics very colored by local views, views which often stem from some stake holder's specific agenda.
Basically, you're talking about an ideal that is unachievable. In other words, going in this direction like it's actually going to work is rather naive.
I'm not advocating anarchy as a means of governing any kind of work (literary or otherwise). Naturally accountability is a requirement for any kind of collaborative work to succeed over time. However I don't think people should confuse accountability with exclusion. All people who so choose may identify themselves directly or via a suitable identity for the purpose of their involvement in a collaborative effort like Wikipedia. Contributing using an identity generates reputation (be it good or bad) and by looking at previous contributions and the value of those, "bad apples" can be ignored. This is in principle not very dissimilar to how slashdot works.
Astute domain experts will under such a system automatically generate "good karma" while still allowing for other people with something of value to contribute to do so.
Regarding your jibe at "Open Source", if you look at the most successful projects it should be clear to you that they do not operate in anarchy but are knitted by groups of core people per project. These people most often accept contributions which are of good quality and which furthers the usefulness of the project delivery and / or the future strategy. As such, the end result may not necessarily be comparable to something like wikipedia, since the former may be governed by the vision of a few (similar to controlled works like a political speech, a survey ordered by a specific multicorp, a report on intelligent design etc.), while the latter aims at spreading knowledge and ideas which inherently always come colored (subjective views on a topic).
Side note: Facts are just claims given a specific status in a particular framework. If the framework was to change, so should the reference for elevating claims to facts. It isn't a static thing we're talking about here, but something volatile... Well I'll stop at that since I don't want the discussion to get off track. Suffice to say that the more minds at a problem, the likelier it is the result will be answers which are less volatile due to the process of arriving at them, due to them having had to work their way through several different frameworks (cultural, ideological, "historical" aside from purely logical).
Has a case against region locking ever reached a court in any country?
I'd imagine the US or at least the EU should have slammed protectionism attempts like these long ago. DVD movies should have provided ample opportunity for making an example out of this free market prevention tactic. Since I still see region coded movie DVDs being sold, I simply have to conclude that the EU has failed miserably in their vigilance to promote free flow of goods. Since the free flow of goods is one of the cornerstones for existance of the EU, this is pretty serious from a European standpoint. It turns the stated goal into a mockery, where one could add subjective exceptions to the statement like "Free flow of goods some goods" or "Free flow of goods unless you pay to get an exception".
In my mind, there can be no logical reason other than bribery for why region coding / locking is still permitted in countries at least trying to pay lip service to the notion of a free market. No wonder a lot of people in Europe view the EU as a big scam, there to serve not it's citizen foremost, but some other stake holder.
Slashdot Mirror
What the T-series really needs is a boost to per-thread performance since it will otherwise remain a specialty processor only suitable for certain workloads.
The T2 core has more than enough parallelism for most apps out there. What isn't appreciated though is that it pushes the server *implementation details* all the way up to the app-developers, which causes them grief when they need to target different hardware or when they utilize "junior" developers. It also causes a lot longer performance tuning phases than on our previous platforms (SPARC and Intel).
This situation is fine for A-list developers but causes major grief for the multitude of companies whose developers don't have experience in massively parallelized systems. Companies in that situation unfortunately are most out there.
Our local SUN.. I mean Oracle drones always point out that our servers are able to handle so much in parallel, but that means squat if we can't meet our SLAs.
If I have a response time SLA of 1 second, then it does me little good if I can service 10 times the number of requests of a competitor's server if each request takes 3 seconds and the competitor's hardware actually allows me to reach the SLA!
Also, if there is any kind of locking going on, the server will more or less halt and whopty doo there goes the parallelism.
We won't be buying another T2 since event the PHBs can read the productivity charts and risk reports handed to them by external consultants. The cost of performance tuning of apps has climbed a lot and cost us a small fortune and continue to do so. Hardest to cope with in this space are legacy and third party apps where hardly anyone dare update the decayed code (or receive funding to do so) or in the latter case can.
However, we're seriously contemplating buying a bunch of Nehalem-EX servers and would perhaps have bought the Power7 if we were an IBM-shop since both those companies "Get It", get what customers as us need in contrast to Oracle.
Advice to Oracle: Add a bunch of cache and allow for higher clock speed in the T3 to really start competing with Intel and IBM. I don't care if you add a thousand more threads if each thread still incur a latency three to four times longer than your competition.
Finally, yes I realize I come across as a sour grape but the amount of time and cost *wasted* as a result of our PHB buying these servers (based on spec-mark figures) without contemplating the intended workload has really put a dent in my department's work atmosphere.
Well C(++) doesn't force you to use new lines at all for most purposes. In the PP's example the exception would be the pre-processor directives.
So, the PP's solution would fit in just 4 lines of code, way fewer than the 60 required for the Python example, but what does that prove w.r.t. either language implementation of the algorithm?
Lines of code is a real stupid measurement for complexity, productivity, maintainability and / or elegance and I really wish people would stop using it like that.
You left out one important fact, that enforceability of patents in the EU is up to each member nation to decide on. This little fact you forgot to mention means that software patents are null and void for the majority of the EU states.
The sole exceptions seem to be the 52rd state of the US, commonly referred to as the UK and to a lesser degree Germany.
Still, both German and UK courts are in a great flux regarding the entire software patent issue so the overall state of patents in the EU is that they are more or less a complete waste of money and time for companies to acquire.
For that reason the PP makes a good point. If Canonical made a European version which ignored the legal minefield currently plaguing the US software market and offered it up to EU citizen with a disclaimer that anyone else may download it at their own peril, then hopefully that would start a debate in the ROW (rest of the world) regarding the damage / benefit balance of software and process patents.
Unfortunately, politicians have neither the insight nor the time to ponder about hypothetical / abstract situations which means the most efficient way to get through to those people is to *show* them something concrete. When the ROW-politicians see their citizen illegally using a distro variant more capable than what is permitted in the home country, a distro which is legal in other countries btw., they will start pondering why the state of affairs is as it is and the damage the current situation inflicts on their society and economy.
It will not be an overnight change but it would most likely be another spark in the kindle to get a sober discussion going in nations which have completely capitulated mentally on the issue.
You could aim at becoming an operational architect, a person who decides what boxes to get, their configuration and amount, network topologies etc. Several colleagues with that focus haven't touched a line of code in years.
Assignment scopes range from small (e.g. supporting an internal application with infrastructure) to large (enterprise wide architecture spanning multiple data centers). The job includes a lot of requirement mgmt. and specifications. Some project management work is also implied (e.g. TCO calcs & comparisons, rollout planning and staffing..)
The pay is very good, but you may have to seek out larger firms where the complexity to be managed justifies having architects with specialization in the functional *or* operational domain. At smaller firms (read less complex environments) the same architect typically performs both parts of the overall design.
However, if you don't already have a lot of experience with boxes, networks, security and lifecycle management of applications then a good start would be to get that first. As someone else mentioned, networks admins, DBAs, server installation etc. are accessible areas for junior people.
An entire additional nation now needs to be able to post information to Slashdot without risk of reprisals from the government or corporate interests. As such I would like to emplore the Slashdot administrators to enable SSL as an alternative to un-encrypted HTTP traffic for reading and posting to this site.
I am fully aware that SSL will increase the resource use of the site, but if you make it a feature that must be enabled in a user's profile, it wouldn't be a default and thus the performance impact should be manageable. As we all know, anything requiring "opt in" will mean only a fraction of the total population will use it.
If you can spare the CPU-cycles, a good service would be something akin to Google's, where you enable SSL for certain (surveiled) IP-ranges where as Google uses it to "i18n" their pages.
Revised and re-voted within just a couple of hours should be pointed out.
The instigators of this constitutional violation had done their homework and were armed with the appropriate fallbacks (contingency tactics) including revised paragraphs and buttons to press to get the political party whips flying so as to eliminate any self-thinking representative from diverging from the party flock. That is, get the slav... ahm. dron... shee.. I mean elected congressional representatives in line for a second vote on a revised legal proposal. Focus was clearly to just get it through without having it properly reviewed and disclosed to the public, thus the prepared amendments.
Everyone but FRA and the party leaders of the ruling council was against this motion including the opposition, the secret service (SÃPA), the legal establishment, the journalists and not to mention the people of Sweden. Still the mission was deftly executed, I must admit, forced through in record time not even breaking a stride when it was rejected in the morning. Well prepared and executed and kudos goes to the evil masterminds of the plan for a "job well executed".
This is a sad day but one which hopefully has opened up a number of important questions regarding the extent to which Sweden claim to be a democracy. Even your average dumbed down person on the street has started to realize some of the implications of STASI-like surveillance. Other questions which have arisen as part of this fiasco is what value having named candidates on the ballot really has, when at the end of the day those people will still be threatened into following the party line regardless of the opinion of those who hand-picked them.
So what now? If you have a need of communicating with people or organizations in Sweden, require of your Swedish counter part to use encryption for ALL communication. For any company having information infrastructure in Sweden, please consider moving it at least as far as to Finland (an adjacent neighbor with as good an Internet infrastructure as Sweeden has) where this kind of mass surveillance is prohibited. The ISP and phone provider TeliaSonera is already in the works of doing this and that should provide a cue for the rest regarding the gravity of this decision.
The government has all but admitted it intends to use this new tool for corporate espionage as well, to get at inside-information regarding interest changes etc. Not a good situation for corporate interests, unless you're the highest bidder and thus gain more inside information than what you loose to you competitors using the same machine.
As to the comments regarding external threats, well that's just pertaining to the analysis phase. The FRA will gather EVERYTHING from the channels / protocols they deem of interest and will LATER analyze that data during the activity defined as signal surveillance for whatever they're missioned to search for in THAT context.
What hasn't been widely discussed is the fact that FRA has a number of interests which can issue orders to them outside "signal surveilance", including the police, the ruling council of the government, other branches of the military as well as certain private interests. These work orders are executed based on the SAME data being intercepted as part of the "surveilance" directive (I.e. all data). In the future more interests will likely be able to get at the data stored by the FRA and given the amount of people having access to this data, when leaks start becoming more prevelant the floodgates simply are no more. (As an example, the road charging system hadn't even been completed before the police came knocking on the door wanting to make use of that info, eventhough not permitted by current law. Think this will be any different?).
PS. The FRA has ILLEGALLY performed surveillance for many years in areas prohibited by law but that's been constrained to signals in the Ether / air. This new legislation will REQUIRE all ISPs to clone their packets and route the cloned traffic to FRA for storage and later analysis. That's a HUGE scope change and the impact is hard to fathom!
Any extension of copyright by the US was most certainly *not* to maintain parity with the EU but to keep control by Mickey Mouse (read general greed).
There is no general duration of copyrighted works in the US because the US copyright system is a mess
If you don't want people to receive signals your portable devices spew into the ether, don't broadcast (I.e. turn the device off).
Simple eh?
Did you read the article?
The system is claimed to be anonymous and I tend to believe that since your personal data (who you are) isn't broadcasted in the operational signal but reside in a database owned by your specific ISP.
The system seems to be tracking unique phones by signal characteristics, thus allowing the phone's path to be plotted but the data delivered is still just "some phone took this route through the monitored area".
And to limit the "DRM" impact, making the GLONASS, Gallileo and GPS systems transmit both CDMA and FDMA signals, receivers supporting the two formats would essentially eliminate the "DRM" fear.
Sure, each individual player (US, EU or Russia/India) could selectively disable their own system's coverage over a region if they wanted, but the receivers would just pick satellites from the remaining systems.
So it seems rather redundant and costly for a system to incorporate regional off-switches when the power to select satellites is still on the receiver's side.
Asynchronous Javascript and XML (or in practice XmlHttpRequests using any kind of text based wire format) are a nice addition and if used as pure decorators on top of sites driven by server side markup generation, then it is a good thing. The issue I and others have is that what "Ajax" seems to also imply when applied by sites in practice is a heavy reliance on a lot of client side DOM manipulations.
A site designed around the notion that as long as "Firefox and IE" can morph the bootstrap HTML page through an infinite number of morphs to something completely different, then the site is good. Those kinds of sites are neigh impossible to use via Wget, perl, lynx or any client not having:
1) A Javascript engine
2) A DOM engine
3) a special variant of A and B combined in such a way that they replicate the same quirks (attributes and behavior) inherent in IE and Mozilla.
So to sum it up, I don't think anyone has anything negative to say about requesting data fragments as an alternative to doing full posts/gets to the server. It's when people are being forced to one of a select few specific applications in order to use the web that irritation starts surfacing.
CPUs, memory and storage are today getting to a price/performance point where it starts making sense for manufacturers to contemplate building consumer devices on top of a scaled down but familiar platform with few or no license fees associated. Chinese manufacturers are I think leading the way in this area atm. and we'll soon see the fruit of that interest.
./ titled "Is this the year Linux will conquer the desktop?".
This is a good thing. Consumers are many and varied and most of them are non-techies. To sell to non-techies you have to really nail the (user interface) experience and lessons learned during the next 5 years will eventually trickle back the desktop domain.
So "Linux on the desktop" will i.m.o. not be something that will happen until Linux is in most of our tiny devices (iphone/ipod clones, nokia phones, portable media centers, wearable GPS devices / personal network hubs and whatever other gadgets of today and tomorrow.. ). So my guess, 5-6 years before we start seeing Linux widening noticeably on the desktop, but at that point the current obstacle holding Linux on the desktop back will have vanished and then it will be the final time we see an article on
How can we as developers know when we are working around an intended limitation and when we are simply adding "value add" to a product?
If I create a multi-segment downloader extension for Internet Explorer, am I then working around an intended limitation in IE and will I get sued or is it perceived as something which enhances IE and thus avoid a lawsuit from Microsoft?
With this action they've really shown that developing for their platform (they own all the base components) is a very risky business for any developer except their own.
Great analysis.
/. journal or pm it to me since I'm very interested in their response.
Unfortunately I'm not a lawyer and as such may have misunderstood the three laws in play here, but according to this interview segment with Eben Moglen, it sounds that his counter argument will be that through the patent deal with Novell, Microsoft knowing full well of the imminent introduction of GPL V3 knowingly extends their patent "protection" as per the provision of GPL V3. Now I agree, his argument doesn't clearly address the main concern of your analysis but to me he seems to rely on one law (copyright law) being able to affect another (patent law) through a third (contract law) and that to me seems to be a very risky position to take if not just very complex for non-lawyers to wrap their minds around.
It would be great if you could send your analysis to FSF in order for them to provide their reasoning around how they plan to tackle the question. If you do, please either add any reply you may receive to your
Can a copyright license control patent provisions? Can aspects of an agreement made (contract law) between a party and another be automatically extended beyond the agreement by actions performed and circumstances surrounding one of the parties for actions and circumstances not covered by the agreement itself? Muddy water this, to be sure.
Well, these european countries (norway, sweden ..) don't have laws which prohibit cracking or circumventing DRM so that's not an issue.
However, these "bans on DRM" encourages hardware manufacturers and distributors of DRM-shielded data for these devices to provide this restriction-less behavior "out of the box" for average Jane to enjoy.
DRM circumvention schemes are still rather technical and requires more work than simply plugging in your ipod and moving the music to your hard drive for enjoyment on your connected Hi-Fi equipment.
Well, judging by other countries it's likely they will retain the levy and still make it illegal to copy stuff.
Taking Sweden as an example, there they tax all storage media (not just "removable" media) with this levy. Actually it's not really a tax (taxing illegal activities is illegal itself), but a state protected fee which a private organization is allowed to collect and without insight into how, distribute parts of the sum to a secret list of copyright holders.
So I guess you have to look forward to being coerced by your government into paying levies (which should exempt you by covering any IP loss right?) and then be put behind bars. It's a hypocrisy, but hey at least it isn't the first in the law book.
It's seldom a good summary is posted on Slashdot. Typically the poster provides some nonsense or poses some arbitrary unrelated question at the same time. I hope more posters take after kdawson in actually reading the article, digesting its contents and add some value to the slashdot readers by exposing the gist of the article.
We've got too much crap-publishing going on and anything and anyone who helps reducing the need for readers to weed through crap is a good thing.
Hmm, I've been reflecting on the responsibilities of manages, PMs and architects for the last few weeks and my conclusion after about 10 year in the business is this (after an initial outline):
Architect
An architect must know the business well, so that the the problems in the domain contexts can be identified and resolved. The person also need to know a wide set of technologies and principles so that a viable technical solution to the business problems can be devised and managed. The most challenging part of an architects job is balancing the various forces/needs/desires/priorities, such as money, features, quality, time as well as factoring in the available skills pool containing the resources (people) who will realize the solution as well as those who will maintain it. Already at the high level PM-like activities are performed such as creating time estimates for each component and dependencies between the component deliverables are created. This is then mapped to a project plan with activities parallelized as needed to fit the current realization constraints (e.g. time vs. cost) which is handed over to a project manager to "execute" (follow up). A person having this role the flack is the solution doesn't meet the requirements.
Project Manager
A person who more or less sticks to and revises a Microsoft Project plan which is created to a large extent either by or together with the architect. The PM also manages the internal bureaucracy when it comes to hiring / staffing teams and ensuring the budget is factored into the realization of the project (e.g. if unforeseen events such as delays or mis-calculations strikes against the budget, a remedy activity must be initiated). The PM is more or less an administrator which is hopefully following some method for project delivery. Saying that a PM is an administrator is not a degenerative statement because every project *needs* at least one administrator to off-load and shield the project team from external turbulence and activities not relating to the progress of the realization of the solution. This is essentially ensuring that the cogs are greased so that the machine works smoothly. A person having this role gets a whopping if a project doesn't meet the dead-line / runs over budget.
Line Manager (which most people tend to refer to when they say "manager")
A person responsible for hiring (and firing) people to the company. Administrates the yearly assessments of employees and helps set individual development plans for the people s/he manages. Is also responsible for distributing the yearly bucket of money for salary increases to the team. Is an echo-chamber for the manager above him/her. First-line managers are the most common and less desirable positions in a company because at that position one has to echo unpopular/stupid decisions "downward" to the people who are actually bringing in money to the company, while trying to manage the outrage from the "floor". Kind of like being stuck between a hammer and an anvil. A person having this job is hired to get flack from upper management or the managed people (all the way up to the exec. who gets hit by the board if the news are bad).
Now, a PM in the tech-industry who has a background as a tech-lead or architect would surely be more desirable than a PM who just graduated "PM-school" or a senior one who has never been part of the actual delivery of technical solutions. For a manager, it would perhaps not be as important as for the PM, but for the people being managed to be able to talk to the manager in situations where they end up under a bad PM for example, the manager would be more likely to understand the problems faced by those people if that mgr. understood the "language" and the set of typical project deliverables. E.g. A specialist who is in a crit-sit project who is over burdened with activities and who has an inexperienced PM might be able to resolve the situation by talking to their manager. In those discussions hands-on experience by the manager would help him/
Nearly as important is the fact that the support contract forces me to upgrade or else Microsoft won't give me support.
Windows 2000 was perfectly fine for me but seeing as it was retired last year I had to move to XP (couldn't afford the extended support contract. Cheaper just getting XP licenses).
We have printers for printing rendered data (text/images) onto papers. Why not have a stone or metal printer which uses a lazer with a bot more watts to etch the information into these durable materials? I concede though that searching these etched medias would be more of a problem, but there has been no system that I know of to this date which were capable of accessing vast amounts of information, packed into a small space other than digital/magnetic/photonic storage (comparatively transient in nature).
After the nuclear winter, the next civilization will simply have to "scan" the recorded knowledge of 21th century from rune stones into computers again.. We did it so why wouldn't they..
PS. I hear the moon is pretty rocky, so perhaps we can use it for something after all.
- Allow me to use my laptop on their network and allow me to connect to my company's network through their corporate network or through a dedicated network they've set up on the side, using a VPN client.
- Allow me to use my laptop on their network, but do not provide me with a network for reaching my company's network.
- Do not allow me to use my laptop on the corporate network, but instead provide a "standard laptop", same as the rest of the workforce is using or a desktop to use.
In the first case, I'm always more productive and cheaper per unit of delivery than the other options, because I've got all my tools setup and on average all software (except perhaps one or two apps) I need to perform my work already licensed. I'm also able to harvest the information, knowledge and assets available through my company's network. Some clients realize the benefit of this and instead regulate which tools I may use and what I may do while connected to their network by making me sign an agreement (along with the standard NDA).In the second and third scenario, clients still require me to sign the same "usage agreement" and NDA, so there really is no benefit for them from a liability standpoint. The risk is the same for all three cases and mitigated the same way (through agreements). On the downside, each unit I deliver costs more because I don't have my toolbox and don't have the software configured for optimal productivity. However I still deliver reasonable value per dollar spent because I am still able to leverage resources from my company by using my own laptop on the side (either through a client provided network or a 3G card).
There is really no benefit that I've witnessed from a client perspective in not allowing me to use my own machine and tools.
Now, the quoted statement of having me use my own machine coupled with the limitation of not having my tools available and also not being able to use the resources available through my company's network for the client's benefit seems like a worse situation for a client than any of the previous ones.
No matter from which angle I try comparing the three typical scenarios with using ACE for the suggested purpose, I fail to see the bottom-line benefits for a company relying on consultants / contractors.
- High availability (reducing the likelihood of packet loss between producer and consumer)
- How discretionary provisioning of packets may be be realized (tagging data for consumers or to a set of consumers in a branch of their proposed network tree model, fanning out from producer to consumers, kind of like an LDAP directory tree).
The thing which bothers me is that the patent is a mix of a "for dummies" guide of a set of concepts which are obvious and have been taught at schools and at various tech-companies for over a decade, coupled with some very broad suggestions of concepts of obvious use. The "innovative" part (read. specifics of the proposed solution rather than "innovation", since it's arguable whether there is any innovation in those details as well) is both hard to isolate and account for perhaps only 3-4% of the patent. The result I fear is that this patent may be used to attack a broad range of people and companies delivering solutions making use of any one (or a set of) the 95+% common knowledge concepts which redundantly were incorporated in this patent application.In my mind, this is a bad patent from the outlook that it looks more like a legal weapon enabling attacks on solutions deviced from public knowledge, while it's not providing enough focus or detail to be used as a blue print for creating a content distribution network (which seemed to be part of the aim). A good patent for patent lawyers and their income, but not so much for society and the economy at large.
I've seen worse, but this one isn't close to qualify being in the upper quadrant of beneficial patents.
Since Sony went "digital" (minidisc) I have mostly associated Sony with obscure audio formats (atrac), lossy transcoding and impossible to get my music off the portable player. With Blueray, the perception of a anti-consumer company is strengthened even more due to all the DRM crap for thri higher fidelty video format.
Piled on top of that your three examples and it's stupendously clear the company has as much allure as does the British kitchen.
Well, I think it depends on the topic. If it's a scientific topic, then the truth can be proven (given the acceptance of a current framework of reasoning and deduction). If a topic on the other hand is more subjective in nature (like history tends to be or social topics) then I disagree that there are absolute facts. Recall the phrase "The victor writes the history". This alone implies there may well be a lot of factual errors in "official" or "published" literary works and in that context I'd not necessarily put more value to [enter-your-favorite-encyclopedia], than on a source where topics may get formed from the combination of multiple angles on the same subject. Examples of this would be topics regarding the "war on terror" or middle east politics which happen to be topics very colored by local views, views which often stem from some stake holder's specific agenda.
I'm not advocating anarchy as a means of governing any kind of work (literary or otherwise). Naturally accountability is a requirement for any kind of collaborative work to succeed over time. However I don't think people should confuse accountability with exclusion. All people who so choose may identify themselves directly or via a suitable identity for the purpose of their involvement in a collaborative effort like Wikipedia. Contributing using an identity generates reputation (be it good or bad) and by looking at previous contributions and the value of those, "bad apples" can be ignored. This is in principle not very dissimilar to how slashdot works.
Astute domain experts will under such a system automatically generate "good karma" while still allowing for other people with something of value to contribute to do so.
Regarding your jibe at "Open Source", if you look at the most successful projects it should be clear to you that they do not operate in anarchy but are knitted by groups of core people per project. These people most often accept contributions which are of good quality and which furthers the usefulness of the project delivery and / or the future strategy. As such, the end result may not necessarily be comparable to something like wikipedia, since the former may be governed by the vision of a few (similar to controlled works like a political speech, a survey ordered by a specific multicorp, a report on intelligent design etc.), while the latter aims at spreading knowledge and ideas which inherently always come colored (subjective views on a topic).
Side note: Facts are just claims given a specific status in a particular framework. If the framework was to change, so should the reference for elevating claims to facts. It isn't a static thing we're talking about here, but something volatile... Well I'll stop at that since I don't want the discussion to get off track. Suffice to say that the more minds at a problem, the likelier it is the result will be answers which are less volatile due to the process of arriving at them, due to them having had to work their way through several different frameworks (cultural, ideological, "historical" aside from purely logical).
Has a case against region locking ever reached a court in any country?
I'd imagine the US or at least the EU should have slammed protectionism attempts like these long ago. DVD movies should have provided ample opportunity for making an example out of this free market prevention tactic. Since I still see region coded movie DVDs being sold, I simply have to conclude that the EU has failed miserably in their vigilance to promote free flow of goods. Since the free flow of goods is one of the cornerstones for existance of the EU, this is pretty serious from a European standpoint. It turns the stated goal into a mockery, where one could add subjective exceptions to the statement like "Free flow of goods some goods" or "Free flow of goods unless you pay to get an exception".
In my mind, there can be no logical reason other than bribery for why region coding / locking is still permitted in countries at least trying to pay lip service to the notion of a free market. No wonder a lot of people in Europe view the EU as a big scam, there to serve not it's citizen foremost, but some other stake holder.