The current president has already generously demonstrated that one can make crucial decisions unhindered by any knowledge. The fact is that the opinions of the sitting president matter, even if they are superstitious. So whose opinions do you like most? This matters, "even" if you yourself don't understand anything about the economy.
For it matters whether people share ideas or prejudices, regardless of the wisdom of them. As even the Financial Times backs Obama, you can expect a wave of optimism among investors and entrepreneurs if he gets elected. The "vibes" about the US economy would be good, and because stock valuations and such are essentially predictions about the future, that alone could give the economy a boost. Emotional imponderables, but strictly rational economy is something for academic textbooks.
For me as an outsider, it seems that Obama's spending plans are marginally more realistic, although they still are a long way from restoring health to the US budget. But Americans are going to have to pay more taxes, regardless of what a candidate may promise to potential voters, to pay off the debts and make room to invest in education and infrastructure. Making plans for the future is nice, but not if you are deeply in the red.
And his ideas for restructuring healthcare just make more sense to me. Don't let dark tales of the high cost of "nationalized" health care scare you. The sad truth is that the current US health care "system" is by far the most expensive on the planet, by a 50% margin over the next runner-up, and more than two times the cost of alternative systems that provide a similar quality of care. There is room to restructure healthcare, greatly improve the quality of services, and save $2000 per American per year.
Perhaps spending $150,000 on her wardrobe makes apparent sense on the short term, but it is rather foolish if you think it over. After considering how much John Edwards was laughed at for his expensive haircut, it is hardly imaginable that any candidate would want to appear in the news with a shopping spree like this. Especially with a financial crisis being in full swing.
It would be stupid to assume that nobody will notice. (Even if it didn't have to be officially disclosed, somebody would leak the hockey mom's expensive taste.) Almost everything a candidate does, including buying shoes, is potentially a news story, and the candidates would be smart to act accordingly. In a case like this, you can be absolutely certain that the press will make a story of it.
Of course Conservatives do grumble that it is unfair for the press to put it in the headlines, but that's just silly. It's politics, people, and public scrutiny, even unfair scrutiny, is part of the game. TV channels and newspapers exist because people choose to read and watch them; it is called free press and a free market. Truman would have said: Get out of the kitchen.
I think the bottom line is: If Palin wasn't prepared to go on stage and tell the world "I just spent US$150,000 on clothes and I am proud of it", she shouldn't have done it at all.
It has its uses. I don't really recommend it, but for some applications Excel will do pretty well. It's a matter of time, scale, and complexity: A spreadsheet is very useful if you have up to a few thousand data points, but your experiment is a one-off or has few repeats, and the data analysis is simple. And sometimes "playing with the data" in Excel gives more insight than doing an advanced statistical analysis in some sophisticated package, based on the wrong assumptions.
This is mainly the case for exploratory lab work, where the detailed experiment setup and layout can and will vary a lot from one experiment to another, from day to day. If you know your away around Excel's addressing functions, it is a very handy and flexible tool to recast the data, do a basic analysis, and plot the results. It isn't very sophisticated, but often it doesn't need to be.
I agree that for scientific purposes Excel leaves a lot to be desired, and there are much better tools. If the bulk of the data becomes large, or the analysis complicated, I will typically use Python or MathCAD. I probably should try R, when I have some time to learn it.
But actually, if you know your way around, Excel can be pretty powerful. I once spent an amusing fifteen minutes with a sales rep who tried to sell his product by listing all the features he thought to be available in his product, but not in Excel -- and I explained to him that they could be all done, and without writing any VBA. Admittedly, you had to do something more than click a button.
I've looked at OpenOffice Calc, but it would certainly take some getting used to and it lacks some features that I find really necessary, even at the most basic level. Contour plots, for example, are hard to do without, even allowing for the fact that Excel's are truly horrible.
This seems to be in the nature of the political system of the USA, if you think about it.
Both large parties are diverse conglomerates, containing people who often disagree significantly on important issues. The big question for the party is where to draw the border of the part affiliation. Who is in, and who is out?
The answer to that is that the optimal popular appeal for a party is 50% plus one vote. Less, and you lose the election. More, and you have increased the internal stress in the party and reduced the size everybody's slice of power, for no real purpose. The art of winning elections is to convince the median voter without alienating the rabid zealots at the other end of the party too much.
So the two parties will always align themselves around the median voter. Close elections are in the nature of the system.
I wonder what their approach to maintenance and service will be. These tasks tends to be rather critical for a database system. I also expect that for many of Oracle's customers, running cost would be the decisive factor, and not purchase cost.
Maintenance costs vary rather wildly from the around 3% of purchase cost per year you expect to be charged for next-business-day-on-site service purely for computer hardware, over the about 10% per year that a supplier of really high-tech stuff will charge you, to the 500% or more per year that an IT department may charge to keep a computer running, maintained, secure, and backed-up. I've seen figures up to 2000%, admittedly as the bloated end result of an apparatchik mentality.
The attractiveness of a standard vendor system may be economies of scale -- if Oracle and HP can sell a lot of them, maintenance on them could be relatively cheap. For people who don't really need a carefully tuned and powerful system, that could be very attractive.
In my experience, an animosity between 'business' and 'tech' can grow from both sides.
The most common cause, I think, is a tendency of 'tech' people to bury themselves entirely in their own comfort zone, to do their IT stuff and not look at anything else. This is only too natural, especially when IT is organized as a separate department at high level, but it tends to result in a IT framework that is often irrelevant to the business and sometimes positively harmful. (Particularly grating on normal people is the Religion of IT Policy, according to which things need to be done purely because they are Policy.) Anyway, responsible IT people should always keep an eye on the overall picture and their mind focused on the company objectives.
But it also frequently happens that the business side dismisses the technical side as mere infrastructure, and refuses to believe that doing the technical side better can actually be of benefit to the business. In the extreme case, if you put twelve managers in a room and feed them enough coffee, they will evidently conclude that the only thing that really matters is management, and everything else can be outsourced. Of course including the IT department. Now, if you are selling sandwiches or coffee, IT probably is mere plumbing and you don't need to care much, but in the financial sector or in anything high-tech neglecting IT is dangerous.
Both hostile trends have a strong tendency to reinforce each other, and they can have a very damaging impact on a company. Ignoring the contribution the technical side can make to the company is probably the least harmful; it will lead to a bog standard tool set but that is going to be damaging only over the longer term. An IT department that is blind to the business impact of its decisions can actually do very serious damage on the short term, because it can very effectively (if unintentionally) sabotage key activities and have a strong negative impact on employee morale.
For the "why and how" of using the Java libraries, I think Sun's own tutorials are a good starting point. They don't cover all technologies, but they provide pretty good coverage of the core set of libraries, complete with sample code. And they are regularly updated, easily accessible and free, which is more than can be said for the typical Java introductory book, which doesn't contain that much additional information.
You are right that the API docs don't provide this, but they are not intended to. In places they do link to the tutorials, although it could be more frequent. Actually I think good documentation is one of Java's strength; immeasurably better than MSDN and quite good compared to, for example, the Python documentation. They lack the elegance of Kernigan & Ritchie, but then Java is a more complicated environment.
But I admit that there are gaps. For example, I don't think Sun's tutorials describe how to define a custom event, although these quickly become useful if you do more than very basic GUI programming. Of course you can always Google for these.
As for topics to start with, it's been a while since I started to learn Java from a C++ background. I agree that the step is not that big, but it's worth devoting some time to the mechanisms of event handling, multi-threading, and the Collections framework. These are too important to skip, and may hold a few surprises.
Advantages for the cultural tourist: Central location, lots of history, great food, great beer. If you want to travel around in Europe, Belgium is a good starting point.
Here too, you will find that in many companies English is the second and often the first language at work. Learning another language may actually be a problem because most people will automatically switch to English anyway when they hear your accent. Most of my English-speaking colleagues say they get very little opportunity to practice their language skills.
For working practices you have to take into account the cultural difference. Belgians generally work shorter hours than Americans do, and have more time off, but that usually means that we compress all our work in relatively short time slots, which can be quite stressful. I guess Belgians get the same amount of work done in a day as Americans -- in reality limited by what people can usefully do in a day before they are exhausted, especially in a creative job. The more relaxed US approach to work is probably a lot healthier. But yes, you get time to travel.
As for living standards, Belgium is rather diverse. Most of the Dutch-speaking north of the country is rich. Some of the regions in the French-speaking south are quite poor and grim, while others are gradually dragging themselves out of a poverty trap. Brussels is a confusing mix of the two. Certainly at the current exchange rate, life here is expensive.
The IT market here is strong, as far as I know: Our IT managers certainly complain about the difficulty of finding qualified people. And they complain that such IT people as they get in, tend to be greedy teenagers with only their class of company car and type of PDA on their mind. Try not to be that type, and you'll probably find a job.
Pitfalls: Many expats prefer to live in the larger cities, where housing is very expensive and the traffic horrible. For two years, buying a house doesn't seem to make much sense, and because Belgians are compulsive house builders and buyers, the rental market is limited. It is probably better to live in a medium-sized city, which is cheaper and often more pleasant to live in. Distances in this country are very short anyway by US standards. A regular train will usually bring you right to the center of the nearest large city in 30 minutes, so why live there?
Taxes will be a shock: If you are to be an employee, then take into account you only about half of your gross wages will actually be paid to you; the rest goes straight to the tax collector. It's not as bad as it sounds, because gross income is high and this will also cover much of your health insurance and pension fund. Nevertheless it is best to be aware of the numerous special tricks that expats can use to reduce their taxes. Regular travel abroad, for example, is highly recommended. You may need an accountant, because nobody understands our fiscal laws.
Traffic may come as a shock too. By US standards, most European drivers are speed-crazy kamikazes, and Belgians must be among the worst. Although I have to say that things have improved a lot over the last ten years.
Many US-owned firms have branches here, and you may use that path to get posted here. However, there will be a limit to what companies are willing to pay to give an employees short-term foreign experience.
Of course. We employees of an US-owned megacorp have promptly been advised to leave our laptops at home if we can, and encrypt everything if we can't.
Our IT department has also suggested that at major conferences and trade shows, the local subsidiary will set up secure systems which we can use instead of our laptops to access e-mail.
The first thing to strive for, I think, is ensuring that code on which two people work still has structural consistency. Programming can be very personal, and aside from emotional aspects such as indentation and formatting, individual programmers sometimes have very individual approaches to problems. Chaos can ensue if, for example, two people on the same project have very different approaches towards exception handling.
One way to ensure consistency is to regularly review each other's code, at least initially. Discuss solutions that are clearly different, and agree on the best one. Alternatively you can buy a book that describes a good standard approach and a set of accepted design patterns for the technology that you use, and agree that you will both stick to that. For some platforms there are also software tools to enforce basic rules.
Now is also the time, if you haven't already done so, to put some effort in design documentation, as opposed to code documentation. What is the architecture of your system, and why? What are the underlying concepts, what design choices did your already make? What is the general principle according to which your code is organized? Those issues tend to matter little as long as you work alone, because you just keep them in your head, but that changes in a team. There you have to make them visible in some way, so that all people can work from the same baseline concepts.
Another thing that should be carefully managed from the start are internal and external dependencies. Many programmers have their own favourite libraries and technologies which they like to use for specific purposes, and they may even be emotionally attached to them, but in a team effort somebody needs to manage the dependencies. You need to keep track of the impact of changes, and prevent the frivolous introduction of additional external dependencies.
Actually, I am or was not the leader of those teams I was referring to. In my experience it matters relatively little who is the 'top banana', as you put it. I have been in team meetings of all kind of composition, and I find that the key to successful teamwork is to have people with the self-confidence and skills to have an effective debate, regardless of whether they are junior or senior.
The necessary condition for reaching the kind of inspirational educational context you describe is that there should be enough room for useful debate and activity by the other team members. If there is at least a core group of people who can debate things, the others can listen, absorb the issues, and try to formulate their own ideas. But a one-against-all arrangement in which a single person sets the pace and the other people merely follow, reduces most of the people to passivity and effectively deprives them of the opportunity to learn. Alternatively, and even worse, it can lead to "holy wars" in which people stubbornly repeat their own point of view without listening to others.
I find that meetings in which I have an effective counterpart to have a fluid debate with, are much more useful for everybody. It brings potential difficulties out in the open, highlights which technical choices are being made, and provides a much more stimulating environment. Yes, when you work on your own you debate the issues with yourself. However, doing so in team meetings might just be considered slightly eccentric. And there is no real replacement for an independent mind with a fresh look.
By the way, I am notoriously dissatisfied with the current situation -- but hiring the right people to remedy the situation is not easy.
I don't agree at all that putting braces on their own lines makes code more clear. My feeling is that the "waste" of vertical space is harmful because it makes it more difficult to get a quick overview of a code block.
I prefer my code to look fairly "dense" in the vertical direction, because I feel that this highlights the structure more clearly. Whitespace is important, but excessive whitespace makes code look scattered and gives you the feeling you have to chase it all over the screen. Most IDEs automatically highlight matching braces anyway.
I agree on using braces even for single statements, where it is not strictly necessary. This is much less confusing and error-prone.
I also increasingly find braces to be syntactically useful all by themselves, i.e. outside the context of control statements. Braces are useful to divide code in blocks that reflect its structure, and can help to delimit the lifetime of temporary variables.
Ah well, sometimes I think I am like that. Often, really. Yes: Actually, I do think that in comparison with everybody in our IT department, I am a better coder, and a better architect, and a better ontologist. Arrogant nuisance? You bet.
So, from a certified jerk's perspective, let me tell what we need to be effective members of a team: An effective sparring partner. Seriously. What makes me a frustrated, tired and emotional team member, is that so many supposed team meetings are only unidirectional exchanges. Sometimes I sit there and watch the other members of the team grinding their way through all the little problems of system integration. Sometimes they sit there and watch me sketch a new ontology, architecture and object model.
Such a grouping of people is not a team. It is a group of people sitting in the same room, but at cross-purposes, because we are not talking at the same level. At best we succeed in confusing each other, at worst only in boring each other. I think they are parasites. They think I am an arrogant jerk. Rate of progress, in that constellation? Nil.
Yet I can work in a team. I've worked for years in highly effective teams, and with success. I can tell you what made all the difference: The presence of equals to debate issues with, so that we could talk each other through the problems and emerge from the session with the feeling that we had defined better solutions. Perhaps we are all arrogant nuisances, but as long as we understand and respect each other we keep each other in check, and can function as effective team members.
The moral of the story, as far as I am concerned: Effective teams need a core group of several people with the same high level of skill. You can put in a few people with less skill, so that they can learn from the others. But you should never try to assemble a team of one wolf and five sheep, because that becomes either a classroom or a dinner opportunity, never a team.
I can't speak for the engineers, but I think a reasonable case could be made that scientific careers are indeed poorly accessible for women. Because they are, generally speaking, not very family friendly: The standard assumption is that young scientists are willing to work long and irregular hours for modest pay and put up with a long series of short-term funding and temporary contracts. Scientific careers are high-effort, high-risk, and even many men feel that this kind of work culture is not very compatible with family life and responsible behaviour towards their children, and abandon academic research for industry jobs.
However, instituting quota for women seems to be very much the wrong answer, and one that is likely to be treated with some contempt by female scientists. However, call me a cynic, I doubt Congress really cares about that. Female scientists are not a large voting block. And the lawyers who dominate the political professions are, in the depths of their soul, probably not convinced that science really matters that much. (Well, certainly not as much as lawyering.) Defining quota seems a typically lawyerly answer to me.
Besides, in the case of the USA, the country doesn't just have a shortage of female scientists but plainly a shortage of scientists, albeit one that is much alleviated by immigration. The real answer is in making scientific careers more attractive. The reason why Congress is not considering this is not difficult to figure out: It would cost money, if only a modest amount, and any results would only be visible after they have left office.
As it happens errors in the CCR5 receptor occur naturally and with a significant frequency, mainly in European populations. This 'delta32' mutation results in a defective receptor, but the people with it are healthy. There is also a drugs on the market, maraviroc, that specifically inhibits CCR5.
People with the CCR5-Delta32 mutation are 'long term non progressors', they carry the virus but don't develop AIDS, probably because the virus is incapable of destroying their immune system.
HIV is actually found in variants that use the CCR5 or CXCR4 receptors as co-receptors to enter cells. Apparently X4-tropic viruses, while deadly, have a limited effect on people with uncompromised immune systems. Most infections are with R5, and X4 strains evolve in the later stages of disease. I don't think anybody as yet understands why, but this is a very reproducible occurrence.
This IT function is going to leave the IT department as we know it today and will migrate into the business unit itself. What this means to you is that you need to know what your firm does, and even more importantly, how it does it.
Yes! At last some people with common sense. I'd hire them straight away.
And no, I'm not joking. I will always prefer a moderately skilled IT guy who has a sense of what the firm needs to achieve and how he can contribute to it, over the most brilliant IT mind that's only interested in maintaining his servers in all their glorious gleaming perfection.
I need the first person. On-site, for direct support and face-to-face discussion of how we can best achieve our goals. If I ever need the second, I'll contract someone in India. People who don't understand our business are a dime to a dozen.
Unfortunately, the reality is different. The reality is that when IT managers "rationalize", they move the boxes to India, fire the people I need in my organization, and "replace" them with people in India who are unable to fill the void, no matter how brilliant they are. I've seen the future, and I don't like it.
Apparently you have not bothered to check up on the actual story. A link was provided.
What angered J&J was the decision of the American Red Cross to license the symbol to other companies for use, in return for a small donation, on the packaging of toenail clippers and tooth brushes, among other things.
The judge apparently feels that this is a legitimate fund raising activity, which may very well be a valid point of law. He also feels that the 19th century agreement between J&J and the ARC governing the use of the red cross symbol is overridden by the charter Congress gave to the ARC, which gives it full control over the use the red cross symbol. That too, may be a legitimate point.
However, if the morals of "commercializing" the Red Cross are what you take issue with, rather than the legal framework, I don't see how J&J can be claimed to the be more at fault than the ARC itself.
For the debate is not about the ARC asking J&J to stop the commercial use of the red cross, something which they have done for over a century. (That, I think, would have been a sensible option.) And certainly not about J&J asking the ARC to stop the non-commercial use of the red cross; that would be patently absurd. It's about the ARC expanding the commercial use of the symbol. Surely, if its commercial use is a bad idea, then the ARC's decision is hardly wise? Toenail clippers are rarely a matter of life and death.
As for the sensitivity of J&J on the issue, I think their concern about trademark confusion is unrealistic but genuine. J&J's brand name is one of the most valuable on the market, and it is understandable that they are jealously protective of it. Whether this was a smart approach is something different.
Vista is probably OK for home use, if you have modern and fairly standard hardware, and run modern, fairly standard and not-too-demanding software. It has a pretty interface and the software on which 95% of home users rely runs well enough.
But I wouldn't touch Vista with a bargepole for our industry environment, for two reasons. One, I already have a fair number of annoying performance hogs on my computer, courtesy of our IT department. (We are still on Windows 2000.) Why would I add another set, courtesy of Microsoft? That would probably bring the efficiency of my computer down from 40% to 20%.
Secondly, and probably more importantly, too many of our suppliers don't want to support it. Their positions is, reasonably enough, that if you spend US$500000 or more on robotics, it seems silly to entrust its control to a system running Vista. That just isn't stable enough; and it is almost unchallenged that Vista is less stable than XP. (Actually I expect to see more a more Linux systems for such purposes.) Or if you spend $50000 on advanced high-performance scientific code, why run it on an OS that wastes power on glitz?
And as the same software often needs to run on our desktops, that rules out Vista for most of our staff too, unless we give everybody two computers. Which sounds like a silly idea, but we do it fairly often for users who need more power than our standard configurations provide, or need to run software that is only available for a specific OS. However, doing it for up to 200 people would be silly.
The general feeling of our experienced IT staff and consultants is that we should migrate to XP some time this year (yes...) and skip Vista, in the hope that Win 7 will be better.
Probably it doesn't consume that much helium. Traditional rigid airships often had to vent lifting gas to keep the pressure differential within limits as the airship heated or climbed; then they had to drop ballast when they returned to low altitude or entered cold weather. The Zeppelin NT (a ten year old design, BTW) is a semi-rigid airship with an internal air-filled ballonet to maintain constant pressure, like a blimp. Also, it relies for only 90% of its lifting capability on the gas, the rest being supplied by the engines, and changes in lift are compensated by power settings instead of by dropping ballast.
The big downsides of the traditional Zeppelin-type rigid airships were the need for huge, expensive ground crews to assist in docking, and their vulnerability to bad weather. Most of the accidents (and there were quite a few) involved airships breaking up or running out of lifting capability in bad weather. Fires were (combat conditions excepted) rare, although dramatized very much by the destruction of the Hindenburg. Of course, for sightseeing trips, bad weather conditions will be avoided.
Well if I invent the cure for AIDs then I can't give it away? And I can't license my drug patent so that it can't be used unless you plan on giving it away.
Well... Let us suppose you're a rich philanthropist, that you are, for example, Bill Gates:-) And you develop a drug or vaccine that will cure AIDS for, say, 70% of the people who use it. I write "rich" because you're probably not going to spend less than US$ 100 million on it, and probably several times that. And I write 70% because that would be a quite high success rate for such a treatment, given the great diversity of HIV strains and the long series of failures of HIV vaccine trials. Most drugs don't work for all people, anyway. Being a rich philanthropist, you decide to give it away for free. To the great joy, I add, of millions of people.
Now assume that somebody else has an idea to improve your vaccine and deliver one that will, say, cure 85% of infected people. But person X is not a rich philanthropist, and cannot afford to give away the improved cure for free. It is probably going to cost X and his company also several hundred million to develop it and get it approved, and they need to be able to promise their investors a return on investment. So can he charge for it?
If you say that he is also under an obligation to give it away for free, then probably he won't develop it at all. With your patent, you can stop him. And without profits in sight, nobody will provide X with the necessary money. That means that 15% fewer people will be cured of HIV infection, and that's about 5 million people at the current size of the pandemic. So you condemned them to death, or at least to lifelong treatment with anti-retroviral drugs.
Alternatively, we could argue that the original inventor must continue to pay for improvements. In other words, you are obliged (morally, if not legally) to pay for the work of X. I daringly presume that you would not be overjoyed by that. We will seriously reduce the number of potential rich philanthropists if we try anything like this.
Or we conclude that, as you gave away your product for free, it is just and fair that the taxpayer pays for any further developments. Actually I would feel that in this case, there is a strong moral argument that the public should fund an improved cure. However, that is a novel method to establish fiscal policy. And perhaps your original cure works well for 97% of Americans, and the new drug would benefit mainly Africans, and it's an election year, and the US Congress is not going to vote new taxes even for a better cure for AIDS.
Some of this is, shall we say, wildly optimistic? -- and the rest wildly pessimistic...
Universities and the NIH traditionally did a lot of the basic research to identify disease mechanisms and drug targets. Most of industry wouldn't touch this work because it is too high-risk (you may not find anything useful in years) so left it to academics. But the gap between identifying a drug target and actually bringing a new drug onto the market still requires five to ten years of work by several hundred people. This is why every time a new scientific discovery on a disease is reported in the news, a warning will be attached that an actual drug is probably several years away.
The work involves actually finding molecules that interact with the drug target, doing all the chemistry to increase their potency and reduce their toxicity and other bad properties, and getting them first through animal trials and then through extensive clinical trials. The success rate of this process is dismally low. It produces piles of documents that used to be truck-filling but nowadays, fortunately, can be submitted to the FDA on DVD.
Only fairly recently have the NIH and universities started to follow the lead of the industry and started their own drug discovery programmes. Opinions on this are divided. Some people are optimistic that the large efforts of the NIH will deliver results, because the NIH and academic labs combined probably have more research resources than industry. Others gloomily point out that traditionally, universities stink at finding the kind of drugs that you would actually be willing to give to a patient (as opposed to something that works in a test tube). Actually, what universities are doing now is trying to rent out shiny new their labs to industry to generate money from them.
And I don't know of any recent case in which a study that clearly stated that a drug was lethal, was not shown to the FDA. That, frankly, would be too high a risk for any company to take. What does happen is that when studies show some statistical indications of risk, people often tend to take an optimistic view of them, by saying that the risk is manageable or not significant. You may call that irresponsible but there are so many red herrings in toxicological studies that people get inured to them. If the drug is put on the market and patients start to die, then problem signalled earlier is interpreted as a clear warning flag --- but hindsight is always 20/20. The harsh reality is that no drug is 100% safe and some people who take it will die. The choice is one between acceptable risk and unacceptable risk, not between safety and danger.
As for banning adverts for prescription drugs from television, I think the USA is one of the few countries where this is allowed. It certainly isn't here.
Another flaw with private pharma research is different companies will not co-operate.
Who says? It happens all the time -- if it is advantageous. Companies develop compounds to market them together, support clinical studies of other companies, license out drugs that they don't want to develop, and exchange knowledge and data on scientific conferences. Not that long ago I actually received training from people who work for the competition. Pharma companies are fairly rational about this.
Packaging two drugs in one pill may be trickier because it actually amounts to help a competitor sell his drug. Companies often will draw the line there. But then, drug formulation is a difficult art too, and putting two drugs in one pill isn't as trivial as it sounds and may be far more expensive than you imagine.
As for universities licensing a drug to a pharmaceutical company in return for "a cut of the profits", they can try — but most universities don't have drugs to offer, only targets and tools, and they tend to have badly inflated expectations of what pharmaceutical companies are able and willing to pay for those. Such negotiations frequently drag out or break down. On the whole, companies far prefer to deal with commercial suppliers.
Interesting. I have the somewhat dubious honour of being put in charge of a similar alignment exercise, mostly after I commented on IT capability and alignment with our business goals in terms that are barely fit to print.
To be fair to our IT people, they understood very well that there was a mismatch between IT capabilities and the needs of the company, and they support the exercise, from low down to the highest management levels -- with perhaps a few exceptions. And they really have a point when they highlight that the user groups have not been making it any easier for them, and I want to strongly support them in that.
But the problem is that I haven't yet been able to get the IT-savvy people at the user side and the IT staff to work together as a team. It's a diverse team of specialists rather than generalists, and all have their own views and expectations. There also is an understandable but unwelcome tendency to use our meetings to release years of pent-up frustrations.
Identifying the super-users and making them do the work wouldn't really help us: The potential super-users all have their own high workload, and they don't want to take on the IT workload as well. What we want (desperately) is to make IT work well enough to shoulder a larger part of the burden, and not take on more work ourselves.
The history of past wars suggests to me that the crucial distinction to make is the one between the enemy means of production and logistics, which on the whole tends to be civilian in nature but nevertheless a valid military target, and the society and population of the enemy as a whole.
Military strategies tend to fail when they don't succeed in making that distinction, for intellectual or technical reasons. In Vietnam, for example, the USA clearly demonstrated a willingness to lay waste to large tracts of land and significantly hurt the civilian population of the North. But the effort was poorly directed, both for political and for technical military reasons, and had a too limited impact on North Vietnam's actual ability to wage war. There was often a brutal mismatch between means and ends, with the means being often too sophisticated for the purpose. For example the destruction of a single Soviet-built truck by an US fighter jet was once calculated to cost, on average, US$188,000.
Something similar happened in World War II. The indiscriminate bombing of German cities by British and American bombers was not only legally and morally questionable, it was an inefficient and mostly counter-productive military tactic. When efforts were finally focused on German fuel production and transport, however, the results were catastrophic for the war machine of the 3rd Reich.
What this means, in my opinion, is that correctly "identifying" targets and non-targets is not the difficult part. The real difficulty lies in knowing what to target. Human tactical leaders are not just trained to select what people to shoot at, they identify the important tactical positions and interpret enemy behaviour.
Relying too much on robots, and removing people from the battlefield, could be a handicap in this regard, because it means that the target selection list is going to be pre-programmed, and the positional goals pre-set. Robots could be tricked into going after the wrong target, or their firmware might contain an obsolete target list. A human enemy might be able to figure this out, and switch tactics accordingly. I bet that a human army can still "get in the decision loop" when confronted with robotic opponents, and stay one step ahead.
What this leads to, of course, is a need for the software of gun-toting robots to be rewritten or re-parametrized on the battlefield to meet changing conditions. In challenging conditions, possibly under fire, certainly under time pressure. How's that for an SQA problem?
Actually, trying to teach the IT department to create software can be the biggest challenge. Among the end users there will always be a few people that can write their own software, and as they are usually single-mindedly pursuing their goal without being distracted by mountains of paperwork, they sometimes do very well and often are more than adequate. They may be sloppy in their technical practice, but often not sloppier than nominally qualified programmers (alas), and at least they understand what the software is supposed to do.
The problems begin when you try to hand off software development and support to an IT department that has 1 programmer, 2 documentation managers, 3 database administrators, 5 testers, 7 security managers, 11 project managers, 13 general managers and 17 generally useless people. The teeth-to-tail ratio of general IT departments trends towards to the truly awful.
Slashdot Mirror
The current president has already generously demonstrated that one can make crucial decisions unhindered by any knowledge. The fact is that the opinions of the sitting president matter, even if they are superstitious. So whose opinions do you like most? This matters, "even" if you yourself don't understand anything about the economy.
For it matters whether people share ideas or prejudices, regardless of the wisdom of them. As even the Financial Times backs Obama, you can expect a wave of optimism among investors and entrepreneurs if he gets elected. The "vibes" about the US economy would be good, and because stock valuations and such are essentially predictions about the future, that alone could give the economy a boost. Emotional imponderables, but strictly rational economy is something for academic textbooks.
For me as an outsider, it seems that Obama's spending plans are marginally more realistic, although they still are a long way from restoring health to the US budget. But Americans are going to have to pay more taxes, regardless of what a candidate may promise to potential voters, to pay off the debts and make room to invest in education and infrastructure. Making plans for the future is nice, but not if you are deeply in the red.
And his ideas for restructuring healthcare just make more sense to me. Don't let dark tales of the high cost of "nationalized" health care scare you. The sad truth is that the current US health care "system" is by far the most expensive on the planet, by a 50% margin over the next runner-up, and more than two times the cost of alternative systems that provide a similar quality of care. There is room to restructure healthcare, greatly improve the quality of services, and save $2000 per American per year.
Perhaps spending $150,000 on her wardrobe makes apparent sense on the short term, but it is rather foolish if you think it over. After considering how much John Edwards was laughed at for his expensive haircut, it is hardly imaginable that any candidate would want to appear in the news with a shopping spree like this. Especially with a financial crisis being in full swing.
It would be stupid to assume that nobody will notice. (Even if it didn't have to be officially disclosed, somebody would leak the hockey mom's expensive taste.) Almost everything a candidate does, including buying shoes, is potentially a news story, and the candidates would be smart to act accordingly. In a case like this, you can be absolutely certain that the press will make a story of it.
Of course Conservatives do grumble that it is unfair for the press to put it in the headlines, but that's just silly. It's politics, people, and public scrutiny, even unfair scrutiny, is part of the game. TV channels and newspapers exist because people choose to read and watch them; it is called free press and a free market. Truman would have said: Get out of the kitchen.
I think the bottom line is: If Palin wasn't prepared to go on stage and tell the world "I just spent US$150,000 on clothes and I am proud of it", she shouldn't have done it at all.
It has its uses. I don't really recommend it, but for some applications Excel will do pretty well. It's a matter of time, scale, and complexity: A spreadsheet is very useful if you have up to a few thousand data points, but your experiment is a one-off or has few repeats, and the data analysis is simple. And sometimes "playing with the data" in Excel gives more insight than doing an advanced statistical analysis in some sophisticated package, based on the wrong assumptions.
This is mainly the case for exploratory lab work, where the detailed experiment setup and layout can and will vary a lot from one experiment to another, from day to day. If you know your away around Excel's addressing functions, it is a very handy and flexible tool to recast the data, do a basic analysis, and plot the results. It isn't very sophisticated, but often it doesn't need to be.
I agree that for scientific purposes Excel leaves a lot to be desired, and there are much better tools. If the bulk of the data becomes large, or the analysis complicated, I will typically use Python or MathCAD. I probably should try R, when I have some time to learn it.
But actually, if you know your way around, Excel can be pretty powerful. I once spent an amusing fifteen minutes with a sales rep who tried to sell his product by listing all the features he thought to be available in his product, but not in Excel -- and I explained to him that they could be all done, and without writing any VBA. Admittedly, you had to do something more than click a button.
I've looked at OpenOffice Calc, but it would certainly take some getting used to and it lacks some features that I find really necessary, even at the most basic level. Contour plots, for example, are hard to do without, even allowing for the fact that Excel's are truly horrible.
This seems to be in the nature of the political system of the USA, if you think about it.
Both large parties are diverse conglomerates, containing people who often disagree significantly on important issues. The big question for the party is where to draw the border of the part affiliation. Who is in, and who is out?
The answer to that is that the optimal popular appeal for a party is 50% plus one vote. Less, and you lose the election. More, and you have increased the internal stress in the party and reduced the size everybody's slice of power, for no real purpose. The art of winning elections is to convince the median voter without alienating the rabid zealots at the other end of the party too much.
So the two parties will always align themselves around the median voter. Close elections are in the nature of the system.
I wonder what their approach to maintenance and service will be. These tasks tends to be rather critical for a database system. I also expect that for many of Oracle's customers, running cost would be the decisive factor, and not purchase cost.
Maintenance costs vary rather wildly from the around 3% of purchase cost per year you expect to be charged for next-business-day-on-site service purely for computer hardware, over the about 10% per year that a supplier of really high-tech stuff will charge you, to the 500% or more per year that an IT department may charge to keep a computer running, maintained, secure, and backed-up. I've seen figures up to 2000%, admittedly as the bloated end result of an apparatchik mentality.
The attractiveness of a standard vendor system may be economies of scale -- if Oracle and HP can sell a lot of them, maintenance on them could be relatively cheap. For people who don't really need a carefully tuned and powerful system, that could be very attractive.
In my experience, an animosity between 'business' and 'tech' can grow from both sides.
The most common cause, I think, is a tendency of 'tech' people to bury themselves entirely in their own comfort zone, to do their IT stuff and not look at anything else. This is only too natural, especially when IT is organized as a separate department at high level, but it tends to result in a IT framework that is often irrelevant to the business and sometimes positively harmful. (Particularly grating on normal people is the Religion of IT Policy, according to which things need to be done purely because they are Policy.) Anyway, responsible IT people should always keep an eye on the overall picture and their mind focused on the company objectives.
But it also frequently happens that the business side dismisses the technical side as mere infrastructure, and refuses to believe that doing the technical side better can actually be of benefit to the business. In the extreme case, if you put twelve managers in a room and feed them enough coffee, they will evidently conclude that the only thing that really matters is management, and everything else can be outsourced. Of course including the IT department. Now, if you are selling sandwiches or coffee, IT probably is mere plumbing and you don't need to care much, but in the financial sector or in anything high-tech neglecting IT is dangerous.
Both hostile trends have a strong tendency to reinforce each other, and they can have a very damaging impact on a company. Ignoring the contribution the technical side can make to the company is probably the least harmful; it will lead to a bog standard tool set but that is going to be damaging only over the longer term. An IT department that is blind to the business impact of its decisions can actually do very serious damage on the short term, because it can very effectively (if unintentionally) sabotage key activities and have a strong negative impact on employee morale.
For the "why and how" of using the Java libraries, I think Sun's own tutorials are a good starting point. They don't cover all technologies, but they provide pretty good coverage of the core set of libraries, complete with sample code. And they are regularly updated, easily accessible and free, which is more than can be said for the typical Java introductory book, which doesn't contain that much additional information.
You are right that the API docs don't provide this, but they are not intended to. In places they do link to the tutorials, although it could be more frequent. Actually I think good documentation is one of Java's strength; immeasurably better than MSDN and quite good compared to, for example, the Python documentation. They lack the elegance of Kernigan & Ritchie, but then Java is a more complicated environment.
But I admit that there are gaps. For example, I don't think Sun's tutorials describe how to define a custom event, although these quickly become useful if you do more than very basic GUI programming. Of course you can always Google for these.
As for topics to start with, it's been a while since I started to learn Java from a C++ background. I agree that the step is not that big, but it's worth devoting some time to the mechanisms of event handling, multi-threading, and the Collections framework. These are too important to skip, and may hold a few surprises.
Advantages for the cultural tourist: Central location, lots of history, great food, great beer. If you want to travel around in Europe, Belgium is a good starting point.
Here too, you will find that in many companies English is the second and often the first language at work. Learning another language may actually be a problem because most people will automatically switch to English anyway when they hear your accent. Most of my English-speaking colleagues say they get very little opportunity to practice their language skills.
For working practices you have to take into account the cultural difference. Belgians generally work shorter hours than Americans do, and have more time off, but that usually means that we compress all our work in relatively short time slots, which can be quite stressful. I guess Belgians get the same amount of work done in a day as Americans -- in reality limited by what people can usefully do in a day before they are exhausted, especially in a creative job. The more relaxed US approach to work is probably a lot healthier. But yes, you get time to travel.
As for living standards, Belgium is rather diverse. Most of the Dutch-speaking north of the country is rich. Some of the regions in the French-speaking south are quite poor and grim, while others are gradually dragging themselves out of a poverty trap. Brussels is a confusing mix of the two. Certainly at the current exchange rate, life here is expensive.
The IT market here is strong, as far as I know: Our IT managers certainly complain about the difficulty of finding qualified people. And they complain that such IT people as they get in, tend to be greedy teenagers with only their class of company car and type of PDA on their mind. Try not to be that type, and you'll probably find a job.
Pitfalls: Many expats prefer to live in the larger cities, where housing is very expensive and the traffic horrible. For two years, buying a house doesn't seem to make much sense, and because Belgians are compulsive house builders and buyers, the rental market is limited. It is probably better to live in a medium-sized city, which is cheaper and often more pleasant to live in. Distances in this country are very short anyway by US standards. A regular train will usually bring you right to the center of the nearest large city in 30 minutes, so why live there?
Taxes will be a shock: If you are to be an employee, then take into account you only about half of your gross wages will actually be paid to you; the rest goes straight to the tax collector. It's not as bad as it sounds, because gross income is high and this will also cover much of your health insurance and pension fund. Nevertheless it is best to be aware of the numerous special tricks that expats can use to reduce their taxes. Regular travel abroad, for example, is highly recommended. You may need an accountant, because nobody understands our fiscal laws.
Traffic may come as a shock too. By US standards, most European drivers are speed-crazy kamikazes, and Belgians must be among the worst. Although I have to say that things have improved a lot over the last ten years.
Many US-owned firms have branches here, and you may use that path to get posted here. However, there will be a limit to what companies are willing to pay to give an employees short-term foreign experience.
Of course. We employees of an US-owned megacorp have promptly been advised to leave our laptops at home if we can, and encrypt everything if we can't.
Our IT department has also suggested that at major conferences and trade shows, the local subsidiary will set up secure systems which we can use instead of our laptops to access e-mail.
The first thing to strive for, I think, is ensuring that code on which two people work still has structural consistency. Programming can be very personal, and aside from emotional aspects such as indentation and formatting, individual programmers sometimes have very individual approaches to problems. Chaos can ensue if, for example, two people on the same project have very different approaches towards exception handling.
One way to ensure consistency is to regularly review each other's code, at least initially. Discuss solutions that are clearly different, and agree on the best one. Alternatively you can buy a book that describes a good standard approach and a set of accepted design patterns for the technology that you use, and agree that you will both stick to that. For some platforms there are also software tools to enforce basic rules.
Now is also the time, if you haven't already done so, to put some effort in design documentation, as opposed to code documentation. What is the architecture of your system, and why? What are the underlying concepts, what design choices did your already make? What is the general principle according to which your code is organized? Those issues tend to matter little as long as you work alone, because you just keep them in your head, but that changes in a team. There you have to make them visible in some way, so that all people can work from the same baseline concepts.
Another thing that should be carefully managed from the start are internal and external dependencies. Many programmers have their own favourite libraries and technologies which they like to use for specific purposes, and they may even be emotionally attached to them, but in a team effort somebody needs to manage the dependencies. You need to keep track of the impact of changes, and prevent the frivolous introduction of additional external dependencies.
Actually, I am or was not the leader of those teams I was referring to. In my experience it matters relatively little who is the 'top banana', as you put it. I have been in team meetings of all kind of composition, and I find that the key to successful teamwork is to have people with the self-confidence and skills to have an effective debate, regardless of whether they are junior or senior.
The necessary condition for reaching the kind of inspirational educational context you describe is that there should be enough room for useful debate and activity by the other team members. If there is at least a core group of people who can debate things, the others can listen, absorb the issues, and try to formulate their own ideas. But a one-against-all arrangement in which a single person sets the pace and the other people merely follow, reduces most of the people to passivity and effectively deprives them of the opportunity to learn. Alternatively, and even worse, it can lead to "holy wars" in which people stubbornly repeat their own point of view without listening to others.
I find that meetings in which I have an effective counterpart to have a fluid debate with, are much more useful for everybody. It brings potential difficulties out in the open, highlights which technical choices are being made, and provides a much more stimulating environment. Yes, when you work on your own you debate the issues with yourself. However, doing so in team meetings might just be considered slightly eccentric. And there is no real replacement for an independent mind with a fresh look.
By the way, I am notoriously dissatisfied with the current situation -- but hiring the right people to remedy the situation is not easy.
I don't agree at all that putting braces on their own lines makes code more clear. My feeling is that the "waste" of vertical space is harmful because it makes it more difficult to get a quick overview of a code block.
I prefer my code to look fairly "dense" in the vertical direction, because I feel that this highlights the structure more clearly. Whitespace is important, but excessive whitespace makes code look scattered and gives you the feeling you have to chase it all over the screen. Most IDEs automatically highlight matching braces anyway.
I agree on using braces even for single statements, where it is not strictly necessary. This is much less confusing and error-prone.
I also increasingly find braces to be syntactically useful all by themselves, i.e. outside the context of control statements. Braces are useful to divide code in blocks that reflect its structure, and can help to delimit the lifetime of temporary variables.
Ah well, sometimes I think I am like that. Often, really. Yes: Actually, I do think that in comparison with everybody in our IT department, I am a better coder, and a better architect, and a better ontologist. Arrogant nuisance? You bet.
So, from a certified jerk's perspective, let me tell what we need to be effective members of a team: An effective sparring partner. Seriously. What makes me a frustrated, tired and emotional team member, is that so many supposed team meetings are only unidirectional exchanges. Sometimes I sit there and watch the other members of the team grinding their way through all the little problems of system integration. Sometimes they sit there and watch me sketch a new ontology, architecture and object model.
Such a grouping of people is not a team. It is a group of people sitting in the same room, but at cross-purposes, because we are not talking at the same level. At best we succeed in confusing each other, at worst only in boring each other. I think they are parasites. They think I am an arrogant jerk. Rate of progress, in that constellation? Nil.
Yet I can work in a team. I've worked for years in highly effective teams, and with success. I can tell you what made all the difference: The presence of equals to debate issues with, so that we could talk each other through the problems and emerge from the session with the feeling that we had defined better solutions. Perhaps we are all arrogant nuisances, but as long as we understand and respect each other we keep each other in check, and can function as effective team members.
The moral of the story, as far as I am concerned: Effective teams need a core group of several people with the same high level of skill. You can put in a few people with less skill, so that they can learn from the others. But you should never try to assemble a team of one wolf and five sheep, because that becomes either a classroom or a dinner opportunity, never a team.
I can't speak for the engineers, but I think a reasonable case could be made that scientific careers are indeed poorly accessible for women. Because they are, generally speaking, not very family friendly: The standard assumption is that young scientists are willing to work long and irregular hours for modest pay and put up with a long series of short-term funding and temporary contracts. Scientific careers are high-effort, high-risk, and even many men feel that this kind of work culture is not very compatible with family life and responsible behaviour towards their children, and abandon academic research for industry jobs.
However, instituting quota for women seems to be very much the wrong answer, and one that is likely to be treated with some contempt by female scientists. However, call me a cynic, I doubt Congress really cares about that. Female scientists are not a large voting block. And the lawyers who dominate the political professions are, in the depths of their soul, probably not convinced that science really matters that much. (Well, certainly not as much as lawyering.) Defining quota seems a typically lawyerly answer to me.
Besides, in the case of the USA, the country doesn't just have a shortage of female scientists but plainly a shortage of scientists, albeit one that is much alleviated by immigration. The real answer is in making scientific careers more attractive. The reason why Congress is not considering this is not difficult to figure out: It would cost money, if only a modest amount, and any results would only be visible after they have left office.
As it happens errors in the CCR5 receptor occur naturally and with a significant frequency, mainly in European populations. This 'delta32' mutation results in a defective receptor, but the people with it are healthy. There is also a drugs on the market, maraviroc, that specifically inhibits CCR5. People with the CCR5-Delta32 mutation are 'long term non progressors', they carry the virus but don't develop AIDS, probably because the virus is incapable of destroying their immune system. HIV is actually found in variants that use the CCR5 or CXCR4 receptors as co-receptors to enter cells. Apparently X4-tropic viruses, while deadly, have a limited effect on people with uncompromised immune systems. Most infections are with R5, and X4 strains evolve in the later stages of disease. I don't think anybody as yet understands why, but this is a very reproducible occurrence.
Yes! At last some people with common sense. I'd hire them straight away.
And no, I'm not joking. I will always prefer a moderately skilled IT guy who has a sense of what the firm needs to achieve and how he can contribute to it, over the most brilliant IT mind that's only interested in maintaining his servers in all their glorious gleaming perfection.
I need the first person. On-site, for direct support and face-to-face discussion of how we can best achieve our goals. If I ever need the second, I'll contract someone in India. People who don't understand our business are a dime to a dozen.
Unfortunately, the reality is different. The reality is that when IT managers "rationalize", they move the boxes to India, fire the people I need in my organization, and "replace" them with people in India who are unable to fill the void, no matter how brilliant they are. I've seen the future, and I don't like it.
Apparently you have not bothered to check up on the actual story. A link was provided.
What angered J&J was the decision of the American Red Cross to license the symbol to other companies for use, in return for a small donation, on the packaging of toenail clippers and tooth brushes, among other things.
The judge apparently feels that this is a legitimate fund raising activity, which may very well be a valid point of law. He also feels that the 19th century agreement between J&J and the ARC governing the use of the red cross symbol is overridden by the charter Congress gave to the ARC, which gives it full control over the use the red cross symbol. That too, may be a legitimate point.
However, if the morals of "commercializing" the Red Cross are what you take issue with, rather than the legal framework, I don't see how J&J can be claimed to the be more at fault than the ARC itself.
For the debate is not about the ARC asking J&J to stop the commercial use of the red cross, something which they have done for over a century. (That, I think, would have been a sensible option.) And certainly not about J&J asking the ARC to stop the non-commercial use of the red cross; that would be patently absurd. It's about the ARC expanding the commercial use of the symbol. Surely, if its commercial use is a bad idea, then the ARC's decision is hardly wise? Toenail clippers are rarely a matter of life and death.
As for the sensitivity of J&J on the issue, I think their concern about trademark confusion is unrealistic but genuine. J&J's brand name is one of the most valuable on the market, and it is understandable that they are jealously protective of it. Whether this was a smart approach is something different.
Vista is probably OK for home use, if you have modern and fairly standard hardware, and run modern, fairly standard and not-too-demanding software. It has a pretty interface and the software on which 95% of home users rely runs well enough.
But I wouldn't touch Vista with a bargepole for our industry environment, for two reasons. One, I already have a fair number of annoying performance hogs on my computer, courtesy of our IT department. (We are still on Windows 2000.) Why would I add another set, courtesy of Microsoft? That would probably bring the efficiency of my computer down from 40% to 20%.
Secondly, and probably more importantly, too many of our suppliers don't want to support it. Their positions is, reasonably enough, that if you spend US$500000 or more on robotics, it seems silly to entrust its control to a system running Vista. That just isn't stable enough; and it is almost unchallenged that Vista is less stable than XP. (Actually I expect to see more a more Linux systems for such purposes.) Or if you spend $50000 on advanced high-performance scientific code, why run it on an OS that wastes power on glitz?
And as the same software often needs to run on our desktops, that rules out Vista for most of our staff too, unless we give everybody two computers. Which sounds like a silly idea, but we do it fairly often for users who need more power than our standard configurations provide, or need to run software that is only available for a specific OS. However, doing it for up to 200 people would be silly.
The general feeling of our experienced IT staff and consultants is that we should migrate to XP some time this year (yes...) and skip Vista, in the hope that Win 7 will be better.
Probably it doesn't consume that much helium. Traditional rigid airships often had to vent lifting gas to keep the pressure differential within limits as the airship heated or climbed; then they had to drop ballast when they returned to low altitude or entered cold weather. The Zeppelin NT (a ten year old design, BTW) is a semi-rigid airship with an internal air-filled ballonet to maintain constant pressure, like a blimp. Also, it relies for only 90% of its lifting capability on the gas, the rest being supplied by the engines, and changes in lift are compensated by power settings instead of by dropping ballast.
The big downsides of the traditional Zeppelin-type rigid airships were the need for huge, expensive ground crews to assist in docking, and their vulnerability to bad weather. Most of the accidents (and there were quite a few) involved airships breaking up or running out of lifting capability in bad weather. Fires were (combat conditions excepted) rare, although dramatized very much by the destruction of the Hindenburg. Of course, for sightseeing trips, bad weather conditions will be avoided.
Well... Let us suppose you're a rich philanthropist, that you are, for example, Bill Gates :-) And you develop a drug or vaccine that will cure AIDS for, say, 70% of the people who use it. I write "rich" because you're probably not going to spend less than US$ 100 million on it, and probably several times that. And I write 70% because that would be a quite high success rate for such a treatment, given the great diversity of HIV strains and the long series of failures of HIV vaccine trials. Most drugs don't work for all people, anyway. Being a rich philanthropist, you decide to give it away for free. To the great joy, I add, of millions of people.
Now assume that somebody else has an idea to improve your vaccine and deliver one that will, say, cure 85% of infected people. But person X is not a rich philanthropist, and cannot afford to give away the improved cure for free. It is probably going to cost X and his company also several hundred million to develop it and get it approved, and they need to be able to promise their investors a return on investment. So can he charge for it?
If you say that he is also under an obligation to give it away for free, then probably he won't develop it at all. With your patent, you can stop him. And without profits in sight, nobody will provide X with the necessary money. That means that 15% fewer people will be cured of HIV infection, and that's about 5 million people at the current size of the pandemic. So you condemned them to death, or at least to lifelong treatment with anti-retroviral drugs.
Alternatively, we could argue that the original inventor must continue to pay for improvements. In other words, you are obliged (morally, if not legally) to pay for the work of X. I daringly presume that you would not be overjoyed by that. We will seriously reduce the number of potential rich philanthropists if we try anything like this.
Or we conclude that, as you gave away your product for free, it is just and fair that the taxpayer pays for any further developments. Actually I would feel that in this case, there is a strong moral argument that the public should fund an improved cure. However, that is a novel method to establish fiscal policy. And perhaps your original cure works well for 97% of Americans, and the new drug would benefit mainly Africans, and it's an election year, and the US Congress is not going to vote new taxes even for a better cure for AIDS.
Some of this is, shall we say, wildly optimistic? -- and the rest wildly pessimistic...
Universities and the NIH traditionally did a lot of the basic research to identify disease mechanisms and drug targets. Most of industry wouldn't touch this work because it is too high-risk (you may not find anything useful in years) so left it to academics. But the gap between identifying a drug target and actually bringing a new drug onto the market still requires five to ten years of work by several hundred people. This is why every time a new scientific discovery on a disease is reported in the news, a warning will be attached that an actual drug is probably several years away.
The work involves actually finding molecules that interact with the drug target, doing all the chemistry to increase their potency and reduce their toxicity and other bad properties, and getting them first through animal trials and then through extensive clinical trials. The success rate of this process is dismally low. It produces piles of documents that used to be truck-filling but nowadays, fortunately, can be submitted to the FDA on DVD.
Only fairly recently have the NIH and universities started to follow the lead of the industry and started their own drug discovery programmes. Opinions on this are divided. Some people are optimistic that the large efforts of the NIH will deliver results, because the NIH and academic labs combined probably have more research resources than industry. Others gloomily point out that traditionally, universities stink at finding the kind of drugs that you would actually be willing to give to a patient (as opposed to something that works in a test tube). Actually, what universities are doing now is trying to rent out shiny new their labs to industry to generate money from them.
And I don't know of any recent case in which a study that clearly stated that a drug was lethal, was not shown to the FDA. That, frankly, would be too high a risk for any company to take. What does happen is that when studies show some statistical indications of risk, people often tend to take an optimistic view of them, by saying that the risk is manageable or not significant. You may call that irresponsible but there are so many red herrings in toxicological studies that people get inured to them. If the drug is put on the market and patients start to die, then problem signalled earlier is interpreted as a clear warning flag --- but hindsight is always 20/20. The harsh reality is that no drug is 100% safe and some people who take it will die. The choice is one between acceptable risk and unacceptable risk, not between safety and danger.
As for banning adverts for prescription drugs from television, I think the USA is one of the few countries where this is allowed. It certainly isn't here.
Who says? It happens all the time -- if it is advantageous. Companies develop compounds to market them together, support clinical studies of other companies, license out drugs that they don't want to develop, and exchange knowledge and data on scientific conferences. Not that long ago I actually received training from people who work for the competition. Pharma companies are fairly rational about this.
Packaging two drugs in one pill may be trickier because it actually amounts to help a competitor sell his drug. Companies often will draw the line there. But then, drug formulation is a difficult art too, and putting two drugs in one pill isn't as trivial as it sounds and may be far more expensive than you imagine.
As for universities licensing a drug to a pharmaceutical company in return for "a cut of the profits", they can try — but most universities don't have drugs to offer, only targets and tools, and they tend to have badly inflated expectations of what pharmaceutical companies are able and willing to pay for those. Such negotiations frequently drag out or break down. On the whole, companies far prefer to deal with commercial suppliers.
Interesting. I have the somewhat dubious honour of being put in charge of a similar alignment exercise, mostly after I commented on IT capability and alignment with our business goals in terms that are barely fit to print.
To be fair to our IT people, they understood very well that there was a mismatch between IT capabilities and the needs of the company, and they support the exercise, from low down to the highest management levels -- with perhaps a few exceptions. And they really have a point when they highlight that the user groups have not been making it any easier for them, and I want to strongly support them in that.
But the problem is that I haven't yet been able to get the IT-savvy people at the user side and the IT staff to work together as a team. It's a diverse team of specialists rather than generalists, and all have their own views and expectations. There also is an understandable but unwelcome tendency to use our meetings to release years of pent-up frustrations.
Identifying the super-users and making them do the work wouldn't really help us: The potential super-users all have their own high workload, and they don't want to take on the IT workload as well. What we want (desperately) is to make IT work well enough to shoulder a larger part of the burden, and not take on more work ourselves.
Can you offer any useful advice?
The history of past wars suggests to me that the crucial distinction to make is the one between the enemy means of production and logistics, which on the whole tends to be civilian in nature but nevertheless a valid military target, and the society and population of the enemy as a whole.
Military strategies tend to fail when they don't succeed in making that distinction, for intellectual or technical reasons. In Vietnam, for example, the USA clearly demonstrated a willingness to lay waste to large tracts of land and significantly hurt the civilian population of the North. But the effort was poorly directed, both for political and for technical military reasons, and had a too limited impact on North Vietnam's actual ability to wage war. There was often a brutal mismatch between means and ends, with the means being often too sophisticated for the purpose. For example the destruction of a single Soviet-built truck by an US fighter jet was once calculated to cost, on average, US$188,000.
Something similar happened in World War II. The indiscriminate bombing of German cities by British and American bombers was not only legally and morally questionable, it was an inefficient and mostly counter-productive military tactic. When efforts were finally focused on German fuel production and transport, however, the results were catastrophic for the war machine of the 3rd Reich.
What this means, in my opinion, is that correctly "identifying" targets and non-targets is not the difficult part. The real difficulty lies in knowing what to target. Human tactical leaders are not just trained to select what people to shoot at, they identify the important tactical positions and interpret enemy behaviour.
Relying too much on robots, and removing people from the battlefield, could be a handicap in this regard, because it means that the target selection list is going to be pre-programmed, and the positional goals pre-set. Robots could be tricked into going after the wrong target, or their firmware might contain an obsolete target list. A human enemy might be able to figure this out, and switch tactics accordingly. I bet that a human army can still "get in the decision loop" when confronted with robotic opponents, and stay one step ahead.
What this leads to, of course, is a need for the software of gun-toting robots to be rewritten or re-parametrized on the battlefield to meet changing conditions. In challenging conditions, possibly under fire, certainly under time pressure. How's that for an SQA problem?
Actually, trying to teach the IT department to create software can be the biggest challenge. Among the end users there will always be a few people that can write their own software, and as they are usually single-mindedly pursuing their goal without being distracted by mountains of paperwork, they sometimes do very well and often are more than adequate. They may be sloppy in their technical practice, but often not sloppier than nominally qualified programmers (alas), and at least they understand what the software is supposed to do.
The problems begin when you try to hand off software development and support to an IT department that has 1 programmer, 2 documentation managers, 3 database administrators, 5 testers, 7 security managers, 11 project managers, 13 general managers and 17 generally useless people. The teeth-to-tail ratio of general IT departments trends towards to the truly awful.