Creating stem cells from adult cells is so far mainly interesting for research purposes. The first hope of researchers is that in the petri dish, culture flask or microtiterplate, stem cells and stem cell derived cell lines may be better research tools than the current cell lines. The cell lines currently used in laboratories are often cancer cell lines and poorly representative of the cells in someones brain or liver. Stem cells may also help us to better understand cell development and what happens when it goes awry.
For therapeutic applications, the first applications may depend on finding drugs that stimulate stem cells to differentiate: It may not be necessary to inject stem cells or cell derived from stem cells, because we may all carry cells with a differentiation potential. For example, regions of the brain seem to contain cells that could potentially differentiate to help people who are suffering brain damage or degeneration.
However, controlling differentiation is complex. While only four factors are sufficient to induce any cell to revert to a stem-cell state, inducing a stem cell to become e.g. a motor neuron is a very complex process that needs to be controlled step by step.
and even if you do, you don't have long before you have to run the attack again.
Not really. Because if people need to change their passwords frequently, they tend to go for stupid changes, such as incrementing a suffix number. I could make a pretty good guess of what some passwords on our systems will be a year from now, even though they are nominally changed every 90 days.
You would be amazed. And frankly, I feel I have too much exposure to IT in the enterprise rather than too little.
Back in the old days, in an organization of a few hundred people, I used to work closely with people in the IT department, and we got along very well. Even if half of the time I found it difficult to understand what they were talking about. But we all knew what we wanted, and that we needed to cooperate, and were successful.
But I have seen IT ever more centralized, first on an European level, then on a global level. And every time the organization became bigger, it also became more concerned with itself and less with users. Users are those nasty people who all insist on having different requirements, when IT "knows" that it is more efficient for all of us to have the same.
I have become a firm believer in the decentralization of IT. Only in small units can you have IT managers who really care and can be engaged in a conversation. The benefits of centralization and standardization are greatly outweighed by its disadvantages.
Windows Vista is a good move because it's been available for some time and they've had enough time to test it out with whatever software they might use.
And crucially, the US military has the clout and money to have its suppliers resolve any bug or incompatibility, even to the point of having software redeveloped if they need to. That is a luxury few of us enjoy. Even today, some of the software packages I need won't run on Vista, or their suppliers recommend against it.
I guess some people don't understand how IT works in organizations with more than a few hundred users.
Unfortunately, I do. Seeing how the IT department just couldn't care less about what the users actually need in an organization with only a few thousand users, I shudder to think how bad it could be in the US military.
Although, to be honest, the best military organizations (and the current US military certainly ranks among them) have always paid great attention to gathering feedback and intelligence from the front line. Some hapless junior-low-cadre victim of the Corporate IT department is unlikely to be able to submit an "unsatisfactory report" on his software installation and have it processed by the command line. A junior officer in the US military probably can.
Easier to use: Maybe, but I find the ribbon a lot slower. My experience is that to some serious job in Excel, you find yourself switching back and forth between tabs all the time. In Word it is not as bad, but I still find it annoying.
The big problem with the Ribbon is that it is insufficiently customizable. In older versions of Office, you could pull frequently used functionality into toolbars, and impose your own ordering and grouping. In comparison the Ribbon only offers the same one-size-fits-nobody solution for everyone. Perhaps it works for the average Office user, but it doesn't work for me.
However, the Ribbon, if an issue at all, is certainly not the biggest problem with the user interface of Office 2007. For example, the biggest downgrade in Excel 2007 functionality is in the pop-up menus for charts. Here you are really slowed down, with functionality scattered around over different pages, and default settings that are inane. And PowerPoint 2007 has a maddingly annoying grid positioning system, which doesn't allow you to put drawn objects where you want them.
I have plenty of clients who would be glad to benefit from letting me use this tool on their systems (and boy would it make my job easier...).
Well, you just lost me as a potential customer:-)
Absolutely the very last thing we need is another tool to expand the IT-run police state. For years now, people have been transferring more and more power to IT in the name of security. It has done great harm to productivity. Please note: Making the work of IT people easier is not synonymous with productivity.
What we do need are OS'es that give users freedom to work as they need to, while providing an adequate --- adequate, not perfect! --- level of security.
MS is not even attempting to go there. We need different solutions, really fast.
The thing could print out reports on a little "credit card receipt" printer, and had a USB port for copying the results to a PC. The USB port was filled with silicone: not validated.
VERY probably it was not the regulatory agency that actually enforced that. The FDA may not be particularly efficient, and they know it, but these days even the FDA prefers that you file a set of DVDs instead of a truckload of paper.
You have to keep an unalterable record of the measurements, but regulations don't say exactly what that means and how you have to do it. You might have to make a print, but typing over the numbers does not make the data any more unalterable than using an USB stick. Actually, to exclude typing errors, it is probable that they had to type them in twice, and then compare the numbers!
A silicone-filled USB port more likely speaks of a bloody-minded compliance team and/or an IT management with bovine excrement between their ears. Those are (massive understatement warning!) not unheard of in pharma, and are largely self-inflicted damage.
If we want to make the pharma companies play nice, we could start by letting the FDA and the EDA (european) cooperate, and say that if something is good enough for the US, it is good enough for Europe, and the other way around.
The European organization is EMEA. It would definitely help, but actually the situation between EMEA and FDA is not that bad: You have to file twice, but at least they will mostly accept the same data. Worse offenders by far are the Japanese agencies, who regularly ask pharma companies to start trials all over again, as if the Japanese people were a different species.
Nothing particularly ironic about it: Drug developers from industry, the academic world, and non-profit organisations rub shoulders on scientific and industry conferences on a fairly regular basis. And there is some mobility between the three communities.
probably too afraid that the alternatives will be more successful than the current patent system, and people will start to wonder why more drugs can't be developed that way.
When it comes to neglected tropical diseases, the patent system has little to do with it. Patents are an issue when a disease afflicts both rich and poor. That sets up friction between pharma companies who want to have profitable pricing and a big price differential between rich and poor countries, and governments who want to push all prices down. However, when it comes to NTDs, usually all the patients are poor, so there is no money, no profits, and often no patents.
It's not entirely true that pharma companies resist any solution. Their preferred solution has been to offer research products with potential benefit, but no potential profit, to non-profit organizations who do the clinical trials. While clinical trials in the industrialized world cost money on a scale that can probably make even Bill Gates blanch, foundations like that of Bill & Melinda can finance much cheaper clinical trials in the third world, if there is a supply of hardy and very determined volunteers. Whether this model will be sufficient in practice remains to be seen.
The problem is in the first research stage, of course: Currently there is no incentive for the companies to do even that. In the future academic research may deliver new drugs, but while universities are now spending lavishly on the necessary equipment for drug discovery, they don't yet have all the skills or experience to make best use of it. And even academic funding agencies often prefer to spend money on the diseases of people they know.
Ah, everybody's favourite sport: Libelling the pharmaceutical industry. Especially funny in combination with the remarkable allegation that nobody seems to question its actions. I guess you spent the last ten years on a trek through the Kalahari?
Having worked in this industry for most of my career, I have to say that I have yet to meet the corrupt pernicious vampires of legend. I have met quite a few idiots, and remarkable number of people who honestly try to combine helping people with running a business. Nobody says it is easy, and nobody says it is perfect. I've heard plenty of harsh words spoken by people within the industry, and at least they knew what they were talking about. I've said some rather harsh things myself.
As it happens, I agree that companies should not be marketing prescription drugs to potential patients. It's a wasteful practice and it encourages even more waste. But as to why they do it...
Too often, patients just want to be given some medicine. And the doctor wants patients to come back. So doctors, if there is no danger involved, are too often willing to oblige patients with a completely useless prescription. Having the doctor "decide what medicine you need" doesn't enter in the equation because you don't need any.
But on the other hand, and this is important, even the best doctor can't actually decide what medicine you need. He or she can only decide what medicine is most likely to help you, an important difference. Even if the diagnosis is free from doubt (it often isn't), people just react to drugs in different ways. A drug that helps one person may trigger a bad allergic reaction in another one. That's why when a patient says "I like treatment X but Y doesn't really work for me", the doctor is likely to listen, and should. In the best case, such statements actually reflect the patient's medical history. I guess that must be pretty rare. Attempts to establish electronic medical records may help, up to a point.
Maybe there will be a day when patients carry their entire genome and epigenomics with them on a card, and doctors will have software that can accurately identify which medicine best suits the patient. Don't hold your breath.
While TNG was infinitely better made, I always much preferred the spirit of TOS -- the 1960s spirit of optimism and curiosity, of exploring space "because it is there". At its core was a belief in the virtues of science (Spock) tempered by morality (Bones).
I always felt that TNG was infected by the pessimism and the pragmatic mindset of the profit-oriented 1990s. Over-compensating for it by going on a search for elaborate political correctness -- "to baldly go where no bald man has gone before" (at least, not without a toupee.)
Besides, TNG had a swaggering arrogance that TOS lacked. We were constantly being told that the Enterprise was the prestigious flagship of the federation, the crew the best of the best, and its captain a tactical genius, a great scientist and the outstanding moral philosopher of his age. Even the insufferable brat on the bridge was presumably someone who we were supposed to admire.
Keep in mind that pharmaceutical companies don't have full freedom in the design of trials. It is, fortunately, a very highly regulated activity.
This trial was probably reviewed by boards of experts and blessed by the regulatory organizations, who approved both the length of the study and the criteria for enrolment. A six-week study may very well produce misleading results, I couldn't comment on that, but it would not be the first time people defined stupid plans with the best of intentions. Typically trials are kept at a minimal length at first to reduce the risk to the volunteers, and then gradually extended in later phases of development.
It is not particularly outrageous in itself that a drug manufacturer should collect a few papers that report favourable data on its products, bundle them with a few adverts and some marketing materials, and hand them out at conferences and trade shows. This happens all the time and it does little harm because you know who the sponsor is, and of course that you should not expect full objectivity.
The problem is in the disguise: Elsevier, a respectable publisher of scientific journals, apparently has a side business "Excerpta Medica", which states on its website that "Excerpta Medica Helps Pharma Companies Fulfill 2009 Pharma Guidelines with Elsevierâ(TM)s Physician and Patient Educational Content." In other words, Excerpta Medica is a marketing organisation that serves pharmaceutical companies. It seems highly unwise for a large scientific publisher to run a side business of this nature, which screams "conflict of interest" pretty loud.
The moral figleaf is provided by the "2009 Pharma Guidelines", issued by the PhRMA. However, the PhRMA is essentially a lobby organization for the pharmaceutical companies. Being a lobbyist is not necessarily evil, and no doubt self-regulation can be a good thing, but nevertheless this figleaf is a bit too small to cover Elsevier's shame: Essentially Excerpta Medica is vowing to obey the moral standards defined by its own customers!
The selling point, of course, is obvious: Elsevier holds copyrights to a vast amount of scientific publications, both journals and books, so it can churn out impressive compilations on demand. Or, as they put it on their website "we can leverage the resources of the worldâ(TM)s largest medical and scientific publisher."
We can only hope that most of these publications will have been peer-reviewed earlier, but Excerpta's website also makes it clear that "authors take full responsibility for the content of their manuscripts" and the editor of the publication is "an outside expert". In other words, Elsevier lends it good name to promotional materials, but declines responsibility for their content.
Developing this sort of infrastructure on a school-by-school basis is incredibly stupid.
Actually, it's called free market thinking.
Even in industry big, centralized IT departments are bureaucratic nightmares that drive the people who have to depend on them insane, not to mention the people unfortunate enough to work in them. They quickly develop into monopolistic organizations that charge their customers anything between 5 to 500 times the market rate for simple items and basic support. Much of that money is then wasted on piles and piles of paperwork. Effective support is replaced by the ritual invocation of regulations. Those of us who routinely deal with Corporate IT departments are familiar with the experience.
And now you want to create a government-run central IT department? You must be joking. Have you noticed the fate of the typical large IT project run on behalf of the government -- also know as a billion pound disaster?
It is much, much better to decentralize and let the schools do their own thing. Yes, they will waste some money; but at least they won't waste hundreds of millions at once on some doomed megaproject. Instead, innovative projects can be started up locally with minimal fuss and bureaucracy, and then they can be adopted by other schools. Some will fail, but others will succeed. And headteachers will (if the unions allow it) actually be able to reward good IT people and sack unsatisfactory ones, which makes for a much healthier relationship and a better IT department.
Decentralize, decentralize, decentralize. Centralizing IT support is a sin you should not commit unless you have a very good excuse indeed. Putting it in the hands of government is a sin that cannot possibly be forgiven.
I'm all for teaching evolution but would someone please explain to me what the issue was with teaching the strengths and weaknesses?
I don't know what exactly was in the Texas list, but in general the problem is that the Creationists have been attacking evolution theory with scant respect for facts, logic, or honesty. As a result, many of their arguments centering on the "weaknesses" of evolution are deeply flawed; to the degree that I would call them intellectual pollution.
A lot of these arguments are based on a simple failure to grasp the concepts of evolution theory, suggesting e.g. that evolution is completely random. Creationist organizations fighting among each other (for there is not such thing as a single theory of creationism) have helpfully compiled lists debunking each others arguments, with titles such as "Arguments we think creationists should NOT use." They are easy to find, though not necessarily correct in their reasoning.
Anyway, there are numerous real bones of contention among evolutionary biologists, and nobody could object against presenting some of these to the students and evaluating the arguments on the different sides of the debates. There is, for example, no general agreement on the cause of the extinction of the dinosaurs, and reviewing the debate on this might make a good classroom exercise.
The biggest problem with security is often that the IT people don't understand what the computers are actually used for. And worse: Don't even want to know. They have converted their IT job into a cargo cult.
They then define security policy as the unilateral invention of the IT department, stressing how to be secure as opposed to how to work securely. Ignoring that the best way to be secure is to pull the plug, of course, as that would put them out of a job as well.
The result is usually an IT policy that conflicts with getting work done, and therefore is undermined by employees at every opportunity. Overall security result: Zero. But lots of mutual loathing and recrimination.
In some fields this is frighteningly common. I've been in debate sessions with a few score of colleagues, most of them working with competing firms, and found them in universal agreement that their IT department was hopeless and they would be better off doing everything themselves. Several of them had already set up their own systems, quick and dirty and probably with pretty poor security. But it worked for them, which is all what mattered to them --- at the time.
The lesson is: Always define your IT policies, security and others, together with the users. Especially the heavier consumers of IT resources and the users with the most skills, for they have the know-how to bust the security systems, and their example will be followed by their peers. Make sure policies are acceptable to everyone and the logic behind them is well understood.
Secondly, make sure to always be there to offer help when someone has a problem that needs to be solved. You want to be part of that solution. And never, never say that it just can't be done.
I understand some of your frustrations, but I have to say that only too often, it doesn't look any better from the user side. Face it, most helpdesks are (thinly) manned by relatively inexperienced people who only took the job because they didn't know yet how tedious and frustrating it is. The service tends to be in line with that.
When you report a problem to them, you probably first have to wait for several days, because there never are enough service people. Then they will take over your computer and make the user watch while they google for the problem and its possible solution. Not necessarily a stupid thing to do, but it tends to undermine confidence. When did you last see a doctor entering your symptoms in Google? At least they don't do it while you watch.
In any large organization, somebody may have had the bright idea of appointing technology or product specialists. This usually means that there are only two people in the IT service team who are allowed to know how to reboot a Linux server. One is now in charge of the company webpage and has been officially banned by his manager from doing any service work on any server whatsoever. The other one is crossing the Sahara on a camel and not expected to return within three weeks, if ever. Users tend to find this mildly frustrating.
And woe to the user who would be so foolish as to have a client-server connectivity problem: This involves at least three people, one for the server, one for the client, and one for the network. Each is apt to report that, whatever the user may say, *his* part of it is working perfectly well -- problem solved. (And besides, they didn't change anything, and if they did nobody saw them do it.) Eternity may pass before they agree on whose fault it is, and what should be done about it.
Nevertheless, the first prize for weeping and gnashing of teeth goes to service organizations who decide that it is much better if they alone decide what hardware and software people should have. Evidently, this makes service and maintenance much easier. And it stands to reason that nobody could or should ever need a tool that is not made by Micro$oft---besides, they were given such a good deal on Vista! No need to ask the users. There is no justification for asking the users, anyway; what do users know, after all? They don't have to maintain the systems.
But absolutely the low point in confidence is reached when the IT service people approach the users and *ask* them to complain about the service. Because this is the only way in which they could hope to force a change in their wretched management.
This may be too optimistic. The situation you describes applies mostly for home users: Their suppliers had few other options than making sure that their software would run on Vista.
But it doesn't really apply to other segments of the market. The Royal Navy recently put Windows in a nuclear submarine -- but they opted for Win 2K and XP. Most companies and large non-commercial buyers don't have applications that are that critical, but the smart ones still avoided Vista and are betting on moving from XP to Win 7. Many specialized software applications continued to run on XP, often the preferred OS for both vendor and customer.
Of course that doesn't necessarily mean that the software can't be run under Vista: Most applications made with recent tools probably will. It does mean, however, that there will have been a minimal effort in optimizing and troubleshooting these applications with Vista. This will now have to be done with Win 7.
SOX was the reaction to a series of big financial scandals, with the Enron affair being the best remembered. But back then, the reaction of many financial experts was to point out the deficiencies of SEC oversight and the weaknesses of the American GAAP accounting rules. The suggested answer was to seriously improve SEC oversight and adopt international IAS standards for accounting.
Instead we got SOX, an only too typically American solution, which puts its faith in forms and auditing. I think at the root of this is the Protestant mindset brought along by the Pilgrim Fathers, which insists that if something is printed on paper it must be true. It is the same mindset that requires travellers to the USA to confirm in writing that they are not terrorists.
And we got -- yes! -- another big financial crisis, caused by lax SEC oversight and creative financial practices. Well, even bigger and badder, if that helps.
The more brains you have on a project generally means the faster it gets done(note I did say Generally).
This is the concept that an IT consultant once described as "using nine women to produce a baby in one month." Beyond a certain critical mass, generally not higher than about seven people, adding more people slows down R&D work. Larger groups can work effectively only if they divide themselves into smaller teams working on well-defined parts of the job.
Even armies have figured this out -- modern armies may be huge and complex organizations, but the smallest tactical unit is a squad of about ten people, much like the Roman contubernium.
Indeed Einstein did not work in complete isolation: Much of the mathematical framework for the theory of relativity was explored by Poincare and Lorentz. And he corresponded about his ideas with others. Nevertheless, theoretical physics at this level is a highly individual activity, because ultimately it is all about thinking and testing concepts in a mathematical framework.
Everybody who has ever written a scientific paper, knows that there is little connection between the number of authors and the number of people who have worked on it.
Often department heads who have barely glanced at a paper are assured of being named as authors. Young researchers who don't even understand the paper, are formally added to the authors list to increase their chances of applying for research funding, much like Royal Navy captions of the Napoleonic era added the names of the sons of their friends to the their ship's books. But the scientific writer who devoted much effort to turning hastily scribbled notes into a readable paper is not even mentioned.
I agree that lonely geniuses are extremely rare. Most people work in the context of their time and rely on the work of others. And besides, many people just have a need to debate their ideas critically with others. Many 'lonely geniuses' kept up an extensive scientific correspondence.
On the other hand, I think the same rule can be applied to science as to programming, i.e. that a good team is at most five to seven people strong. In larger teams communication breaks down quickly and people work at cross-purposes. Adding more people makes discovery slower, not faster. Large institutions may work as long as they provide a framework in which small, effective teams can work without creating too much administrative overhead.
A factor that contributes to the disappearance of the 'lonely genius' is the increasing specialization of scientists. The geniuses of the 17th and 18th century were not 'universal men' any more, but they still had a wide scope of interest and knowledge. These days individuals often have rather narrow knowledge, and teams need to be assemble of people from different fields to make good progress. (It's interesting to look at the recruitment lists of modern biotechnology centers.)
This raises the question whether there is "a limit of specialization", at which a team that is small enough to be effective is also too small to have a sufficiently wide range of knowledge. At which point scientific discovery must slow down to a crawl.
I think the biggest task you have is to translate between the business side and the technical side. In many poor workplaces, the problem is that the programmers don't understand the business purpose they are supposed to serve, and the business people don't understand the technical limitations and practicalities of the software system. As the manager, your job entails putting business goals in understandable language for the programmers, and explaining to business managers what the programmers are doing.
This can be a surprisingly though job, because the two groups of people often talk very different languages, and sometimes even show mutual contempt. But if you don't do it properly, you will see software design done by business managers, and/or business strategy made by programmers -- neither is a particularly good idea.
Your second major task is to stimulate your team to learn and improve, and to give them the opportunity to do so. Your programmers may be 'seasoned', but I don't know whether that means that they are very experienced or smart, or just that they are very experienced and stuck in mind-numbing routine. Anyway, there is always more to learn and good programmers really like to do so. You should protect your team against being so overloaded that there isn't any time to explore new things, and guide them in the direction of exploring something potentially useful.
Chu must have reasonable political skills, as he the director of the Berkeley Lab, an organization with 4000 people and a budget of half a billion. The management of a scientific organization of this nature is usually quite challenging, if only because many of the people employed by it are (necessarily) independent-minded and headstrong. There is more back-stabbing in academic labs than in Washington DC.
Putting a scientist in charge of energy policy is a good idea. A factually justified, realistic energy policy is urgently needed.
Besides, during the last few years people in the public research departments have been demoralized by a political leadership that made it clearly felt that it couldn't care less about scientific data and factual reality. The DoE needs a leader who has the confidence of its staff. Chu could be that leader.
Excellent post, but I would just like to add a few more caveats. As you probably know, but most people certainly don't, HIV occurs in a wide range of variant strains, which use either CD4 and CXCR4 or CD4 and CCR5. This is a property called "tropism" and HIV strains are classified as X4 or R5 tropic.
But actually, from the data I have seen, few viruses exhibit a really pure tropism. There are a lot of dual-tropic X4R5 virus strains that have some flexibility to use CXCR4 or CCR5 as the opportunity offers itself; this is not surprising, as one variant must be capable of evolving into the other. Also, all patients carry diverse virus populations, because HIV is so sloppy in replicating itself, and a patient may well have 99.9% of R5-tropic viruses and 0.1% of X4-tropic viruses. (0.1% is about the limit of what can be detected with current, and very expensive, methods.)
This is cause for concern. A treatment that blocks the replication of R5-tropic viruses may well favour the replication of X4 strains. There already are indications that this happens in some patients on maraviroc treatment. Driving a virus population to become entirely X4 tropic would probably not be advisable, as there is reason to believe that these strains do more damage. The X4-tropic HIV strains are generally associated with the late stages of infection and the development of AIDS.
Therefore I doubt that anyone is going to advocated bone marrow transplants as a way to treat HIV. The risk is just too big, because this form of treatment is (almost) irreversible. Treatment with maravoric should always be preceded by tropism testing, and can be stopped if it doesn't work.
Finally, 2 years of undetectable viral load in absence of ARV treatment is an impressive result, but IMHO it is still too early to call the patient cured. He may well still have proviral DNA in his cells.
The choice facing your company is not the one between having IT support and not having IT support. Nobody in their right mind will seriously ask that question. The choice for your company is between having you on their payroll as a full-time employee, or signing a contract with an IT service company, perhaps one located in India, to do all the IT support work.
As a customer of IT services, I would say that the great value of having an on-site person take care of IT infrastructure is quite simply that you can talk with him. I hope that you talk to other employees, that you listen to their wishes and demands, that you try to find solutions that work well for your company, and that you have people's trust and confidence. This is very important, but difficult to measure. On the long term it ensures that any IT solutions are the right size and the right cost for the business needs, but that is not immediately obvious.
I know from experience that it is extremely important to have someone on-site who actually understands (and shares) the goals of the business. Someone you can have a face-to-face conversation with. You don't really understand the value of that unless you have been forced to miss it. (Of course you can still contract out some of the work.)
The bad news is that this is very difficult to measure. Well, perhaps you could measure the raised blood pressure of workers who have to make do without a local IT contact, and in extreme cases the loss of talented employees who won't tolerate bad IT support. But it will be hard to sell that case to management. Many managers will blithely ignore human factors (don't ask me why) and reduce IT support to a purely technical, almost robotic matter which can be farmed out to India without difficulty. The cost of this kind of policy to the company can be considerable, but it is mostly a hidden cost which doesn't show at all on the budget.
Anyway, my advice, for what it is worth: Don't focus on how much money you save for the company by installing a server or managing desktops. Somebody else is always willing to do it cheaper. Instead, try to get a statement from other employees on how important it is for them to have a good support person they can discuss problems with, and get solutions. And pray that your management has a soul.
Slashdot Mirror
Creating stem cells from adult cells is so far mainly interesting for research purposes. The first hope of researchers is that in the petri dish, culture flask or microtiterplate, stem cells and stem cell derived cell lines may be better research tools than the current cell lines. The cell lines currently used in laboratories are often cancer cell lines and poorly representative of the cells in someones brain or liver. Stem cells may also help us to better understand cell development and what happens when it goes awry.
For therapeutic applications, the first applications may depend on finding drugs that stimulate stem cells to differentiate: It may not be necessary to inject stem cells or cell derived from stem cells, because we may all carry cells with a differentiation potential. For example, regions of the brain seem to contain cells that could potentially differentiate to help people who are suffering brain damage or degeneration.
However, controlling differentiation is complex. While only four factors are sufficient to induce any cell to revert to a stem-cell state, inducing a stem cell to become e.g. a motor neuron is a very complex process that needs to be controlled step by step.
and even if you do, you don't have long before you have to run the attack again.
Not really. Because if people need to change their passwords frequently, they tend to go for stupid changes, such as incrementing a suffix number. I could make a pretty good guess of what some passwords on our systems will be a year from now, even though they are nominally changed every 90 days.
You would be amazed. And frankly, I feel I have too much exposure to IT in the enterprise rather than too little.
Back in the old days, in an organization of a few hundred people, I used to work closely with people in the IT department, and we got along very well. Even if half of the time I found it difficult to understand what they were talking about. But we all knew what we wanted, and that we needed to cooperate, and were successful.
But I have seen IT ever more centralized, first on an European level, then on a global level. And every time the organization became bigger, it also became more concerned with itself and less with users. Users are those nasty people who all insist on having different requirements, when IT "knows" that it is more efficient for all of us to have the same.
I have become a firm believer in the decentralization of IT. Only in small units can you have IT managers who really care and can be engaged in a conversation. The benefits of centralization and standardization are greatly outweighed by its disadvantages.
Windows Vista is a good move because it's been available for some time and they've had enough time to test it out with whatever software they might use.
And crucially, the US military has the clout and money to have its suppliers resolve any bug or incompatibility, even to the point of having software redeveloped if they need to. That is a luxury few of us enjoy. Even today, some of the software packages I need won't run on Vista, or their suppliers recommend against it.
I guess some people don't understand how IT works in organizations with more than a few hundred users.
Unfortunately, I do. Seeing how the IT department just couldn't care less about what the users actually need in an organization with only a few thousand users, I shudder to think how bad it could be in the US military.
Although, to be honest, the best military organizations (and the current US military certainly ranks among them) have always paid great attention to gathering feedback and intelligence from the front line. Some hapless junior-low-cadre victim of the Corporate IT department is unlikely to be able to submit an "unsatisfactory report" on his software installation and have it processed by the command line. A junior officer in the US military probably can.
Easier to use: Maybe, but I find the ribbon a lot slower. My experience is that to some serious job in Excel, you find yourself switching back and forth between tabs all the time. In Word it is not as bad, but I still find it annoying.
The big problem with the Ribbon is that it is insufficiently customizable. In older versions of Office, you could pull frequently used functionality into toolbars, and impose your own ordering and grouping. In comparison the Ribbon only offers the same one-size-fits-nobody solution for everyone. Perhaps it works for the average Office user, but it doesn't work for me.
However, the Ribbon, if an issue at all, is certainly not the biggest problem with the user interface of Office 2007. For example, the biggest downgrade in Excel 2007 functionality is in the pop-up menus for charts. Here you are really slowed down, with functionality scattered around over different pages, and default settings that are inane. And PowerPoint 2007 has a maddingly annoying grid positioning system, which doesn't allow you to put drawn objects where you want them.
I have plenty of clients who would be glad to benefit from letting me use this tool on their systems (and boy would it make my job easier...).
Well, you just lost me as a potential customer :-)
Absolutely the very last thing we need is another tool to expand the IT-run police state. For years now, people have been transferring more and more power to IT in the name of security. It has done great harm to productivity. Please note: Making the work of IT people easier is not synonymous with productivity.
What we do need are OS'es that give users freedom to work as they need to, while providing an adequate --- adequate, not perfect! --- level of security.
MS is not even attempting to go there. We need different solutions, really fast.
The thing could print out reports on a little "credit card receipt" printer, and had a USB port for copying the results to a PC. The USB port was filled with silicone: not validated.
VERY probably it was not the regulatory agency that actually enforced that. The FDA may not be particularly efficient, and they know it, but these days even the FDA prefers that you file a set of DVDs instead of a truckload of paper.
You have to keep an unalterable record of the measurements, but regulations don't say exactly what that means and how you have to do it. You might have to make a print, but typing over the numbers does not make the data any more unalterable than using an USB stick. Actually, to exclude typing errors, it is probable that they had to type them in twice, and then compare the numbers!
A silicone-filled USB port more likely speaks of a bloody-minded compliance team and/or an IT management with bovine excrement between their ears. Those are (massive understatement warning!) not unheard of in pharma, and are largely self-inflicted damage.
If we want to make the pharma companies play nice, we could start by letting the FDA and the EDA (european) cooperate, and say that if something is good enough for the US, it is good enough for Europe, and the other way around.
The European organization is EMEA. It would definitely help, but actually the situation between EMEA and FDA is not that bad: You have to file twice, but at least they will mostly accept the same data. Worse offenders by far are the Japanese agencies, who regularly ask pharma companies to start trials all over again, as if the Japanese people were a different species.
(Ironically, I got that through ScienceDirect).
Nothing particularly ironic about it: Drug developers from industry, the academic world, and non-profit organisations rub shoulders on scientific and industry conferences on a fairly regular basis. And there is some mobility between the three communities.
probably too afraid that the alternatives will be more successful than the current patent system, and people will start to wonder why more drugs can't be developed that way.
When it comes to neglected tropical diseases, the patent system has little to do with it. Patents are an issue when a disease afflicts both rich and poor. That sets up friction between pharma companies who want to have profitable pricing and a big price differential between rich and poor countries, and governments who want to push all prices down. However, when it comes to NTDs, usually all the patients are poor, so there is no money, no profits, and often no patents.
It's not entirely true that pharma companies resist any solution. Their preferred solution has been to offer research products with potential benefit, but no potential profit, to non-profit organizations who do the clinical trials. While clinical trials in the industrialized world cost money on a scale that can probably make even Bill Gates blanch, foundations like that of Bill & Melinda can finance much cheaper clinical trials in the third world, if there is a supply of hardy and very determined volunteers. Whether this model will be sufficient in practice remains to be seen.
The problem is in the first research stage, of course: Currently there is no incentive for the companies to do even that. In the future academic research may deliver new drugs, but while universities are now spending lavishly on the necessary equipment for drug discovery, they don't yet have all the skills or experience to make best use of it. And even academic funding agencies often prefer to spend money on the diseases of people they know.
Ah, everybody's favourite sport: Libelling the pharmaceutical industry. Especially funny in combination with the remarkable allegation that nobody seems to question its actions. I guess you spent the last ten years on a trek through the Kalahari?
Having worked in this industry for most of my career, I have to say that I have yet to meet the corrupt pernicious vampires of legend. I have met quite a few idiots, and remarkable number of people who honestly try to combine helping people with running a business. Nobody says it is easy, and nobody says it is perfect. I've heard plenty of harsh words spoken by people within the industry, and at least they knew what they were talking about. I've said some rather harsh things myself.
As it happens, I agree that companies should not be marketing prescription drugs to potential patients. It's a wasteful practice and it encourages even more waste. But as to why they do it...
Too often, patients just want to be given some medicine. And the doctor wants patients to come back. So doctors, if there is no danger involved, are too often willing to oblige patients with a completely useless prescription. Having the doctor "decide what medicine you need" doesn't enter in the equation because you don't need any.
But on the other hand, and this is important, even the best doctor can't actually decide what medicine you need. He or she can only decide what medicine is most likely to help you, an important difference. Even if the diagnosis is free from doubt (it often isn't), people just react to drugs in different ways. A drug that helps one person may trigger a bad allergic reaction in another one. That's why when a patient says "I like treatment X but Y doesn't really work for me", the doctor is likely to listen, and should. In the best case, such statements actually reflect the patient's medical history. I guess that must be pretty rare. Attempts to establish electronic medical records may help, up to a point.
Maybe there will be a day when patients carry their entire genome and epigenomics with them on a card, and doctors will have software that can accurately identify which medicine best suits the patient. Don't hold your breath.
While TNG was infinitely better made, I always much preferred the spirit of TOS -- the 1960s spirit of optimism and curiosity, of exploring space "because it is there". At its core was a belief in the virtues of science (Spock) tempered by morality (Bones).
I always felt that TNG was infected by the pessimism and the pragmatic mindset of the profit-oriented 1990s. Over-compensating for it by going on a search for elaborate political correctness -- "to baldly go where no bald man has gone before" (at least, not without a toupee.)
Besides, TNG had a swaggering arrogance that TOS lacked. We were constantly being told that the Enterprise was the prestigious flagship of the federation, the crew the best of the best, and its captain a tactical genius, a great scientist and the outstanding moral philosopher of his age. Even the insufferable brat on the bridge was presumably someone who we were supposed to admire.
Keep in mind that pharmaceutical companies don't have full freedom in the design of trials. It is, fortunately, a very highly regulated activity.
This trial was probably reviewed by boards of experts and blessed by the regulatory organizations, who approved both the length of the study and the criteria for enrolment. A six-week study may very well produce misleading results, I couldn't comment on that, but it would not be the first time people defined stupid plans with the best of intentions. Typically trials are kept at a minimal length at first to reduce the risk to the volunteers, and then gradually extended in later phases of development.
It is not particularly outrageous in itself that a drug manufacturer should collect a few papers that report favourable data on its products, bundle them with a few adverts and some marketing materials, and hand them out at conferences and trade shows. This happens all the time and it does little harm because you know who the sponsor is, and of course that you should not expect full objectivity.
The problem is in the disguise: Elsevier, a respectable publisher of scientific journals, apparently has a side business "Excerpta Medica", which states on its website that "Excerpta Medica Helps Pharma Companies Fulfill 2009 Pharma Guidelines with Elsevierâ(TM)s Physician and Patient Educational Content." In other words, Excerpta Medica is a marketing organisation that serves pharmaceutical companies. It seems highly unwise for a large scientific publisher to run a side business of this nature, which screams "conflict of interest" pretty loud.
The moral figleaf is provided by the "2009 Pharma Guidelines", issued by the PhRMA. However, the PhRMA is essentially a lobby organization for the pharmaceutical companies. Being a lobbyist is not necessarily evil, and no doubt self-regulation can be a good thing, but nevertheless this figleaf is a bit too small to cover Elsevier's shame: Essentially Excerpta Medica is vowing to obey the moral standards defined by its own customers!
The selling point, of course, is obvious: Elsevier holds copyrights to a vast amount of scientific publications, both journals and books, so it can churn out impressive compilations on demand. Or, as they put it on their website "we can leverage the resources of the worldâ(TM)s largest medical and scientific publisher."
We can only hope that most of these publications will have been peer-reviewed earlier, but Excerpta's website also makes it clear that "authors take full responsibility for the content of their manuscripts" and the editor of the publication is "an outside expert". In other words, Elsevier lends it good name to promotional materials, but declines responsibility for their content.
Developing this sort of infrastructure on a school-by-school basis is incredibly stupid.
Actually, it's called free market thinking.
Even in industry big, centralized IT departments are bureaucratic nightmares that drive the people who have to depend on them insane, not to mention the people unfortunate enough to work in them. They quickly develop into monopolistic organizations that charge their customers anything between 5 to 500 times the market rate for simple items and basic support. Much of that money is then wasted on piles and piles of paperwork. Effective support is replaced by the ritual invocation of regulations. Those of us who routinely deal with Corporate IT departments are familiar with the experience.
And now you want to create a government-run central IT department? You must be joking. Have you noticed the fate of the typical large IT project run on behalf of the government -- also know as a billion pound disaster?
It is much, much better to decentralize and let the schools do their own thing. Yes, they will waste some money; but at least they won't waste hundreds of millions at once on some doomed megaproject. Instead, innovative projects can be started up locally with minimal fuss and bureaucracy, and then they can be adopted by other schools. Some will fail, but others will succeed. And headteachers will (if the unions allow it) actually be able to reward good IT people and sack unsatisfactory ones, which makes for a much healthier relationship and a better IT department.
Decentralize, decentralize, decentralize. Centralizing IT support is a sin you should not commit unless you have a very good excuse indeed. Putting it in the hands of government is a sin that cannot possibly be forgiven.
I'm all for teaching evolution but would someone please explain to me what the issue was with teaching the strengths and weaknesses?
I don't know what exactly was in the Texas list, but in general the problem is that the Creationists have been attacking evolution theory with scant respect for facts, logic, or honesty. As a result, many of their arguments centering on the "weaknesses" of evolution are deeply flawed; to the degree that I would call them intellectual pollution.
A lot of these arguments are based on a simple failure to grasp the concepts of evolution theory, suggesting e.g. that evolution is completely random. Creationist organizations fighting among each other (for there is not such thing as a single theory of creationism) have helpfully compiled lists debunking each others arguments, with titles such as "Arguments we think creationists should NOT use." They are easy to find, though not necessarily correct in their reasoning.
Anyway, there are numerous real bones of contention among evolutionary biologists, and nobody could object against presenting some of these to the students and evaluating the arguments on the different sides of the debates. There is, for example, no general agreement on the cause of the extinction of the dinosaurs, and reviewing the debate on this might make a good classroom exercise.
The biggest problem with security is often that the IT people don't understand what the computers are actually used for. And worse: Don't even want to know. They have converted their IT job into a cargo cult.
They then define security policy as the unilateral invention of the IT department, stressing how to be secure as opposed to how to work securely. Ignoring that the best way to be secure is to pull the plug, of course, as that would put them out of a job as well.
The result is usually an IT policy that conflicts with getting work done, and therefore is undermined by employees at every opportunity. Overall security result: Zero. But lots of mutual loathing and recrimination.
In some fields this is frighteningly common. I've been in debate sessions with a few score of colleagues, most of them working with competing firms, and found them in universal agreement that their IT department was hopeless and they would be better off doing everything themselves. Several of them had already set up their own systems, quick and dirty and probably with pretty poor security. But it worked for them, which is all what mattered to them --- at the time.
The lesson is: Always define your IT policies, security and others, together with the users. Especially the heavier consumers of IT resources and the users with the most skills, for they have the know-how to bust the security systems, and their example will be followed by their peers. Make sure policies are acceptable to everyone and the logic behind them is well understood.
Secondly, make sure to always be there to offer help when someone has a problem that needs to be solved. You want to be part of that solution. And never, never say that it just can't be done.
I understand some of your frustrations, but I have to say that only too often, it doesn't look any better from the user side. Face it, most helpdesks are (thinly) manned by relatively inexperienced people who only took the job because they didn't know yet how tedious and frustrating it is. The service tends to be in line with that.
When you report a problem to them, you probably first have to wait for several days, because there never are enough service people. Then they will take over your computer and make the user watch while they google for the problem and its possible solution. Not necessarily a stupid thing to do, but it tends to undermine confidence. When did you last see a doctor entering your symptoms in Google? At least they don't do it while you watch.
In any large organization, somebody may have had the bright idea of appointing technology or product specialists. This usually means that there are only two people in the IT service team who are allowed to know how to reboot a Linux server. One is now in charge of the company webpage and has been officially banned by his manager from doing any service work on any server whatsoever. The other one is crossing the Sahara on a camel and not expected to return within three weeks, if ever. Users tend to find this mildly frustrating.
And woe to the user who would be so foolish as to have a client-server connectivity problem: This involves at least three people, one for the server, one for the client, and one for the network. Each is apt to report that, whatever the user may say, *his* part of it is working perfectly well -- problem solved. (And besides, they didn't change anything, and if they did nobody saw them do it.) Eternity may pass before they agree on whose fault it is, and what should be done about it.
Nevertheless, the first prize for weeping and gnashing of teeth goes to service organizations who decide that it is much better if they alone decide what hardware and software people should have. Evidently, this makes service and maintenance much easier. And it stands to reason that nobody could or should ever need a tool that is not made by Micro$oft---besides, they were given such a good deal on Vista! No need to ask the users. There is no justification for asking the users, anyway; what do users know, after all? They don't have to maintain the systems.
But absolutely the low point in confidence is reached when the IT service people approach the users and *ask* them to complain about the service. Because this is the only way in which they could hope to force a change in their wretched management.
This may be too optimistic. The situation you describes applies mostly for home users: Their suppliers had few other options than making sure that their software would run on Vista.
But it doesn't really apply to other segments of the market. The Royal Navy recently put Windows in a nuclear submarine -- but they opted for Win 2K and XP. Most companies and large non-commercial buyers don't have applications that are that critical, but the smart ones still avoided Vista and are betting on moving from XP to Win 7. Many specialized software applications continued to run on XP, often the preferred OS for both vendor and customer.
Of course that doesn't necessarily mean that the software can't be run under Vista: Most applications made with recent tools probably will. It does mean, however, that there will have been a minimal effort in optimizing and troubleshooting these applications with Vista. This will now have to be done with Win 7.
SOX was the reaction to a series of big financial scandals, with the Enron affair being the best remembered. But back then, the reaction of many financial experts was to point out the deficiencies of SEC oversight and the weaknesses of the American GAAP accounting rules. The suggested answer was to seriously improve SEC oversight and adopt international IAS standards for accounting.
Instead we got SOX, an only too typically American solution, which puts its faith in forms and auditing. I think at the root of this is the Protestant mindset brought along by the Pilgrim Fathers, which insists that if something is printed on paper it must be true. It is the same mindset that requires travellers to the USA to confirm in writing that they are not terrorists.
And we got -- yes! -- another big financial crisis, caused by lax SEC oversight and creative financial practices. Well, even bigger and badder, if that helps.
Some people never learn...
The more brains you have on a project generally means the faster it gets done(note I did say Generally).
This is the concept that an IT consultant once described as "using nine women to produce a baby in one month." Beyond a certain critical mass, generally not higher than about seven people, adding more people slows down R&D work. Larger groups can work effectively only if they divide themselves into smaller teams working on well-defined parts of the job.
Even armies have figured this out -- modern armies may be huge and complex organizations, but the smallest tactical unit is a squad of about ten people, much like the Roman contubernium.
Indeed Einstein did not work in complete isolation: Much of the mathematical framework for the theory of relativity was explored by Poincare and Lorentz. And he corresponded about his ideas with others. Nevertheless, theoretical physics at this level is a highly individual activity, because ultimately it is all about thinking and testing concepts in a mathematical framework.
Everybody who has ever written a scientific paper, knows that there is little connection between the number of authors and the number of people who have worked on it.
Often department heads who have barely glanced at a paper are assured of being named as authors. Young researchers who don't even understand the paper, are formally added to the authors list to increase their chances of applying for research funding, much like Royal Navy captions of the Napoleonic era added the names of the sons of their friends to the their ship's books. But the scientific writer who devoted much effort to turning hastily scribbled notes into a readable paper is not even mentioned.
I agree that lonely geniuses are extremely rare. Most people work in the context of their time and rely on the work of others. And besides, many people just have a need to debate their ideas critically with others. Many 'lonely geniuses' kept up an extensive scientific correspondence.
On the other hand, I think the same rule can be applied to science as to programming, i.e. that a good team is at most five to seven people strong. In larger teams communication breaks down quickly and people work at cross-purposes. Adding more people makes discovery slower, not faster. Large institutions may work as long as they provide a framework in which small, effective teams can work without creating too much administrative overhead.
A factor that contributes to the disappearance of the 'lonely genius' is the increasing specialization of scientists. The geniuses of the 17th and 18th century were not 'universal men' any more, but they still had a wide scope of interest and knowledge. These days individuals often have rather narrow knowledge, and teams need to be assemble of people from different fields to make good progress. (It's interesting to look at the recruitment lists of modern biotechnology centers.)
This raises the question whether there is "a limit of specialization", at which a team that is small enough to be effective is also too small to have a sufficiently wide range of knowledge. At which point scientific discovery must slow down to a crawl.
I think the biggest task you have is to translate between the business side and the technical side. In many poor workplaces, the problem is that the programmers don't understand the business purpose they are supposed to serve, and the business people don't understand the technical limitations and practicalities of the software system. As the manager, your job entails putting business goals in understandable language for the programmers, and explaining to business managers what the programmers are doing.
This can be a surprisingly though job, because the two groups of people often talk very different languages, and sometimes even show mutual contempt. But if you don't do it properly, you will see software design done by business managers, and/or business strategy made by programmers -- neither is a particularly good idea.
Your second major task is to stimulate your team to learn and improve, and to give them the opportunity to do so. Your programmers may be 'seasoned', but I don't know whether that means that they are very experienced or smart, or just that they are very experienced and stuck in mind-numbing routine. Anyway, there is always more to learn and good programmers really like to do so. You should protect your team against being so overloaded that there isn't any time to explore new things, and guide them in the direction of exploring something potentially useful.
Chu must have reasonable political skills, as he the director of the Berkeley Lab, an organization with 4000 people and a budget of half a billion. The management of a scientific organization of this nature is usually quite challenging, if only because many of the people employed by it are (necessarily) independent-minded and headstrong. There is more back-stabbing in academic labs than in Washington DC.
Putting a scientist in charge of energy policy is a good idea. A factually justified, realistic energy policy is urgently needed.
Besides, during the last few years people in the public research departments have been demoralized by a political leadership that made it clearly felt that it couldn't care less about scientific data and factual reality. The DoE needs a leader who has the confidence of its staff. Chu could be that leader.
Excellent post, but I would just like to add a few more caveats. As you probably know, but most people certainly don't, HIV occurs in a wide range of variant strains, which use either CD4 and CXCR4 or CD4 and CCR5. This is a property called "tropism" and HIV strains are classified as X4 or R5 tropic.
But actually, from the data I have seen, few viruses exhibit a really pure tropism. There are a lot of dual-tropic X4R5 virus strains that have some flexibility to use CXCR4 or CCR5 as the opportunity offers itself; this is not surprising, as one variant must be capable of evolving into the other. Also, all patients carry diverse virus populations, because HIV is so sloppy in replicating itself, and a patient may well have 99.9% of R5-tropic viruses and 0.1% of X4-tropic viruses. (0.1% is about the limit of what can be detected with current, and very expensive, methods.)
This is cause for concern. A treatment that blocks the replication of R5-tropic viruses may well favour the replication of X4 strains. There already are indications that this happens in some patients on maraviroc treatment. Driving a virus population to become entirely X4 tropic would probably not be advisable, as there is reason to believe that these strains do more damage. The X4-tropic HIV strains are generally associated with the late stages of infection and the development of AIDS.
Therefore I doubt that anyone is going to advocated bone marrow transplants as a way to treat HIV. The risk is just too big, because this form of treatment is (almost) irreversible. Treatment with maravoric should always be preceded by tropism testing, and can be stopped if it doesn't work.
Finally, 2 years of undetectable viral load in absence of ARV treatment is an impressive result, but IMHO it is still too early to call the patient cured. He may well still have proviral DNA in his cells.
The choice facing your company is not the one between having IT support and not having IT support. Nobody in their right mind will seriously ask that question. The choice for your company is between having you on their payroll as a full-time employee, or signing a contract with an IT service company, perhaps one located in India, to do all the IT support work.
As a customer of IT services, I would say that the great value of having an on-site person take care of IT infrastructure is quite simply that you can talk with him. I hope that you talk to other employees, that you listen to their wishes and demands, that you try to find solutions that work well for your company, and that you have people's trust and confidence. This is very important, but difficult to measure. On the long term it ensures that any IT solutions are the right size and the right cost for the business needs, but that is not immediately obvious.
I know from experience that it is extremely important to have someone on-site who actually understands (and shares) the goals of the business. Someone you can have a face-to-face conversation with. You don't really understand the value of that unless you have been forced to miss it. (Of course you can still contract out some of the work.)
The bad news is that this is very difficult to measure. Well, perhaps you could measure the raised blood pressure of workers who have to make do without a local IT contact, and in extreme cases the loss of talented employees who won't tolerate bad IT support. But it will be hard to sell that case to management. Many managers will blithely ignore human factors (don't ask me why) and reduce IT support to a purely technical, almost robotic matter which can be farmed out to India without difficulty. The cost of this kind of policy to the company can be considerable, but it is mostly a hidden cost which doesn't show at all on the budget.
Anyway, my advice, for what it is worth: Don't focus on how much money you save for the company by installing a server or managing desktops. Somebody else is always willing to do it cheaper. Instead, try to get a statement from other employees on how important it is for them to have a good support person they can discuss problems with, and get solutions. And pray that your management has a soul.