Three things:

Even in universities, where wages compared to the private sector are usually lower, IT workers get paid more than double other typical staff. Wage depression is not a big problem right now as far as I can tell. IT workers may not make as much as they think they should make, but that is a different issue. When average IT income starts to approach median household income (instead of being in the top quintile), then we might have something to talk about.

Few industries don't require you to establish yourself and work your way up fresh out of college. Even medical students have to go through residency hell before they can get their first permanent position somewhere. In some places moving up is quick, but starting at a low wage fresh out of college is more normal than abnormal. IT has not had this, for the most part, until recently, so count yourselves lucky, but it can't last forever.

When computing and automation started playing a significant role in the workforce, this exact same conversation happened with respect to the then factory workers and office workers. Did all office or factory work disappear forcing everybody to work slave wages? No, the job landscape changed requiring fewer, but more highly skilled people. Some people will be out of a job. They will have to retrain or change fields. This is what happens. It may take some time, but ultimately those workers end up employed in new areas and growth in other industries occurs as a result.

Yes, but the CRISPR system can be designed to work precisely with a single species, because the targeting sequence can use non-homologous regions of genes that are similar between species. So in your horizontal gene transfer case it would die out after the transfer event into a new species. Another potential safeguard is to put the CRISPR system in a different locus from the mutation, so that horizontal gene transfer events would be very unlikely to transfer both functions into another species.

The entire point behind the method is to not have to do this. You make one genetically engineered organism that then breeds passing on the desired trait, only in such a way that inheritance is biased toward the desired trait so that it isn't lost by "dilution" into the gene pool.

Yes, but the elegance (and limitation) of this system is that it requires sexual reproduction to work. So if the two organisms don't inter-breed, the traits cannot be passed.

Except they didn't say "entire ecosystem". They said "throughout the ecosystem," meaning engineered organism deliberately and purposefully breed with non-modified organisms thus creating a large-scale change in the ecosystem.

This is a it more elegant and controlled, in that it basically just suppresses reversion back to wild type after a mutation has occurred. Nothing else, no need to crest a bazillion untargeted copies all over the place. The process of gene editing (not new) becomes cleaner, which is something greatly needed.

Summary is an excerpt of an article highlighting some potential use of technology developed by George Church's lab at Harvard (and others). It is actually some pretty incredible stuff. Church's first published the adaption of the CRISPR system to gene editing in eukaryotes a few years ago. Basically, it works like this. CRISPR is a bacterial defense system where an enzyme (endonuclease) is directed to cut a specific DNA sequence by it's directly adjacent targeting sequence. Bacteria use this to protect themselves from viruses. When a virus tries to insert itself into the genome of a bacterium, CRISPR will cleave that sequence (if the bacterium has the appropriate targeting system) and subsequent DNA repair processes will occur that will excise the viral sequence. You can think of it as a pseudo-immunity system for bacteria against viruses. Like other DNA sequences, CRISPR sequences can be transferred between bacteria in a population allowing for broad-ranging resistance to viral infection to occur within a bacterial community.

The innovation by Church's group is to put the CRISPR system in eukaryotes. Introducing modified genes by homologous recombination has been around for a long time, but the problem with most eukaryotes is they have multiple copies of each chromosome. So a modification in one copy will get diluted out over several rounds of replication. By including the CRISPR system in the mutation that targets the original gene, a mechanism is supplied to allow a modified gene to quickly spread throughout the population. This makes genetic modification of eukaryotes much more efficient and easier to control.

Now, while safely applies in a laboratory system, the ecological consequences of using such a system in a natural setting are unclear. This is the purpose of the article: to raise some of the issues and possibilities to begin a discussion about how such a system might be used safely and what sort of regulations may need to be put into place. The article does quite a good job of illustrating some scenarios. Here is what I consider the meat of it, but of course other scenarios exist as well.

Just want to put that out there so that a somewhat productive conversation can hopefully happen here.

It's really amazing how much revisionist rubbish gets spouted on Slashdot and treated as gospel. Companies have always since forever viewed their employees from the perspective of a cost-benefit analysis. That is no different now than it was in the 50s or earlier. Why do you think we have things like unions, labor laws, OSHA regulations, minimum wage, etc? When a company seeks to cut costs they will, plain and simple.

The reason companies were willing to train workers in the past was because they had to, not because they were somehow more benevolent toward their employees. Factory and office jobs were on the upswing. People coming out of school, not necessarily high school, didn't have the skills they needed, so they trained them. This, by the way, was a minimum wage job, and factories were installed in small rural towns for a reason.

Now it is different. We are no longer talking about minimum wage factory jobs. We are talking about highly skilled $100k+/yr + benefits tech jobs. It really shouldn't be surprising to anyone why a company wouldn't want to invest in a year of downtime in training, especially when people hop jobs in 2-3 yrs. Yes, a PHP developer can probably become a Perl developer rather quickly, but a DBA cannot become a Perl developer as easily. And why should a company be forced to pay for that anyway? If they can't hire who they need locally, they should be able to look elsewhere. When a company wants to hire fresh grads, how many of those are Americans? Some, but not many.

Tech is experiencing right now what every other field has experienced in the past. They have been shielded from it until fairly recently, but it can't be avoided anymore. The tech workers that can adapt, of their own volition, will probably not have trouble finding decent jobs. The ones that whine and complain because great high paying jobs aren't just handed to them fresh out of college will struggle a bit. Welcome to the rest of the world gentleman. It is not going to go away.

No, Red Hat 6 series used kernel 2.2 and libc6, and Corel definitely ran on it, I remember doing it. Don't know what the problem with Debian was, but it may have been a Wine issue.

No, where did I ever say that? I'm only disputing the claim that you have to get rid of tenure because it doesn't do anything but protect bad teachers. And the subclaim (not yours) that the only way to deal with bad teachers is to fire them.

I agree with everything else you said. I still despise standardized tests, though. Mostly because they fail to distinguish between students who have actually learned something and students who just studied for the test. I do also insist, especially at the high school level, that the home environment and support from the parents is critical for effective learning. In AP classes, you don't see this because students, for the most part, don't take AP classes unless they are motivated to learn. In other classes, not so much. If a teacher doesn't know how to or can't succesfully engage these students, do they still deserve to be fired?

I would suggest avoiding legal remedies, as it is a blunt instrument that people understandably resist when they perceive it as against their interests. Instead, work with the local districts and teachers unions. Come up with effective solutions that both support. When this happens at a sufficient scale, more will follow.

And yet, a tenured teacher can also be like my AP Calculus teacher who put together the best curriculum I have ever been through. So maybe tenure isn't really the problem. I don't know anything about your AP Physics teacher. Was he a passionate physics teacher ever? Was he an english teacher that got promoted to physics? How much of this can be avoided by looking at hiring/evaluation practices and creating an environment where teachers are encouraged and supported to put effort into their classes? My AP Calculus teacher retired early. She loved to teach, but after some 30 yrs, the district bullshit was just too much, and at the time the administration was putting a lot of pressure on senior teachers to retire (because they wanted to save money).

I agree the teacher unions could be a lot more cooperative. But then teacher unions and administrators have had a long history of animosity towards each other. They don't trust each other. Has anybody asked teacher unions to develop an evaluation method that they would support? That would be a good place to start. If the feeling is that somebody is coming in from the outside with a lot of demands, no actual teaching experience, and no consideration for complexities arising from learning disabilities, language barriers, or socioeconomic status, I think they are right to be skeptical.

I agree. This is a good example. Teachers can't teach something they don't know.

I agree. But honestly, my first reaction to seeing a result like that is that there must be something wrong with the test, not the teacher. I don't know how someone just forgets half of their knowledge or how a teacher can induce someone to forget half of their knowledge. More likely, they intensively studied in an ineffective way for the first test. They knew enough to pass the test and then immediately forgot everything. They could have done the same thing for the second test, but then is that a useful evaluation? How much does that student actually know? Another thing to consider, is regurgitating vocabulary an effective way of measuring a student's mastery of the English language? How about prose, composition, grammar? Can they actually use all those words that they know, or just recite definitions?

I note the difference in firing rate. And yet there is little to no evidence that private schools offer a better education or have better outcomes than public schools.

Uh, sure, but is that because of tenure? I doubt it. Are experienced teachers blocking good ideas and protecting themselves from younger teachers with tenure? No, not likely. Experienced teachers are protecting themselves from administrators. The younger teachers don't have this protection, unfortunately.

That is what I said above, but what would make this discussion more interesting would be some numbers. Exactly how many bad teachers are there? As a percentage of total teachers? And how many of those bad teachers have been identified as needing to be removed, but are being blocked by tenure? In my experience, "bad teacher" is often "teacher I don't like" which is another reason tenure exists.

Why is that funny? NCLB didn't come from teachers. This has been a battle between teachers and administrators for a long time as well, since before NCLB. I'll grant that there are plenty of teachers that just do what they are told, but the exceptional ones are the ones that want to do more.

Political influence happens in a lot of different ways. Curriculum decisions, text books, required coursework, etc. I agree that it is rare for specific teachers to be targeted, but any teacher that wants to work against political influence will be vulnerable.

All evaluations are subjective. Just because you can put a number on something doesn't mean it is helpful or meaningful. I don't give a rats ass about standardized test scores. It is fairly trivial to teach people how to take tests. And sure enough, all of the 1600 SAT score people took test prep courses. It says absolutely nothing about actual knowledge or critical thinking skills.

Teachers unions have supported and called for effective evaluations for a long time. Test scores are easy to get and process, but are not very effective. Tell me about efforts to do something beyond the bare minimum for evaluations and we will have something to talk about. It doesn't have to be classroom evaluations. It could be periodic interviews and continuing ed courses, for example. I don't know. Test scores can even be a part of it, but they can't be the only metric.

That is a pretty dire picture of the classroom that you painted. It makes me wonder where you went to school, because that was not my experience at all. Tenure abuses at the primary/secondary school level (Uni tenure is a different beast entirely) might have been there, but I didn't see them. Bad teachers were there for sure, but that is why I propose looking at the evaluation system for teachers, rather than tenure, which is a bit of a red herring.

Just my personal pet theory, but I think it is because school administration is often a stepping stone to higher political careers on the school board, city council, etc. Education is really the least of their concerns. They are more interested in putting on a good face for the political elite, newspapers, activist groups, etc. And they often have only token teaching experience. Their training and their aspirations are to be administrators, not teachers, so they don't easily empathize with the task and problems of teachers.

Tenure is a mixed bag. Yes, it can protect bad teachers, but it also...
      1) Protects experienced senior teachers. You might not think this is important, but guess what? Older, experienced teachers are generally more expensive and have more political influence. Hip new administrator comes in, wants to to change things up, slim down the budget. Get rid of the older teachers first beacuse the younger are cheaper and easier to control.
      2) Protects good teachers. You know the ones that actually teach and care about education, and don't just give A's to everyone for showing up and sitting at their desk. Actual teaching and enforcing academic standards tends to upset certain kinds of parents. Administrators don't like vocal and upset parents.
      3) Protects teachers that push against the administration. Not teaching to the test, enriching the curriculum, doing what might be considered risky things by some ( lab experiments, field trips, etc). Administration often doesn't want this, because it creates headaches for them, but teachers want it because it enhances the education of their students.
      4) In areas with strong influence by outside political groups, protects teachers that teach controversial subjects. Science vs. creationism is one example, but certainly not the only one. History, economics, literature, art...all of these can have controversial topics. Of course, we don't really teach these anymore, but that is a different topic.

Whether or not tenure exists and how it is granted is really missing the point. If you want to improve the quality of teachers, we need to be looking at the evaluation systems that are in place, whether they exist, and why they may or may not be working. Most teachers simply are never evaluated ever, or they are evaluated in completely useless ways. Address that, and then maybe we can deal more easily with underperforming teachers, adjusting the tenure rules as necessary but keeping its major benefits.