Somewhat OT, but Thanks for pointing out the bit about the surfactants, which I've suspected is part of the "magic" for a while. Bought some cheap glyphosate weed killer and it didn't work nearly as well as the Roundup, and that's the only difference I could find on the label. FWIW, I think when I added just a little bit of dish detergent to the off brand stuff and gently mixed, it worked a bit better afterwards.
Not only is there "bureaucratic and political interference", but in all likelihood the financing set aside for decommissioning has been "borrowed against for decades. Sort of like surplus funds in Social Security. But we can't possibly make the poor corps operating the reactors actually come up with the money they were supposed to save for the purpose, because that would be bad for business. Nope, I'm not at all cynical...
Glad to hear you find Researchgate useful for your work.
Around here, though, someone set up a Researchgate account and added a couple of publications to their profile. Researchgate then took the "initiative" to send all coauthors invitations to join. Instead of a somewhat spammy message to the effect of "so-and-so has joined Researchgate; we think you might want to check us out", the message was a much more sleazy "so-and-so has invited you to join him/her on Researchgate". Just want the/. crowd to be aware of the sort of tactics Researchgate is using to try to build their member list.
Sounds simple, but for some of us dropping all mail to/from Yahoo groups isn't feasible. Of six mailing lists that I receive messages from semi-regularly for work related purposes, three of them are on Yahoo groups. All are specific to niche software packages and for two of the three at least represent THE primary source of "support" for those packages. Taking your suggestion wouldn't necessarily be career suicide, but could hurt me.
Having said that, I receive those mailing lists at my work address and not through my Yahoo email account, so the DMARC issue isn't likely to affect me too much. I could probably move my personal mail from the latter, but a) inertia and b) I find the newest incarnations of Gmail not enough better than the abomination the Yahoo mail has become to force the issue.
you should be getting rid of any potential vector for badness (any software, particularly software that is known to touch the internet) altogether.
While you have a point, how do you suggest data get moved on and off that old PC? SFTP/SCP touch the internet. USB sticks are a potential attack vector. We can restrict access to only computers on the LAN that lives behind the firewall, but our other computers have to be able to get outside in order for us to do work, which means that our LAN could in theory get compromised.
Seriously, why don't I have mod points when a comment like this comes up. Seriously, +5 Insightful
I work at a university. Around here, lots of investigators have aging but perfectly serviceable equipment of all flavors. Some of it is tied to XP (or Win9x, or DOS) because the software to run the equipment hasn't been updated to run under a newer OS. Some of the equipment is one off or made by a company that's been out of business for years, or there is a newer and fancier instrument so they won't upgrade software for the old stuff and there's no budget to spend >$100k to upgrade a piece of equipment just so it can run newer software.
This isn't limited to Windows, either. The NMR spectrometer I'm chiefly responsible for has parts (including the acquisition computer) that are nearly 16 years old, and the software we use to control it was released in ~2006. That version of the software will only run under RHEL 4.0 with certain specific PC hardware and drivers. When the Dell PC died ~2 years ago, I could either find a similar box online ($350 delivered) OR I could upgrade 1. the computer (cheapest supported workstation was ~$2500) and 2. buy a license for a newer version of the control software (~$10k). That's not the newest software, which only works with newer hardware.
Fortunately, at least our IT and CompSec people are willing to begrudgingly accept the status quo. They are "mandating" that nearly all Windows boxes get update to at least Win 7 by next Monday. However, for those computers that can't, they are willing to accept keeping them offline as much as possible and behind a strong firewall plus using common sense security measures (i.e. not running normal tasks as admin).
Of course there was always the risk that you might be delayed, which of course meant your dinner might burn. I haven't seen this feature in a while...
Any relatively higher end oven has this feature. Mine does - I can set the oven to turn on at the time and temperature I want and then turn off after a set period of time as well. If you're delayed by a lot, maybe it will be cold again when you get home. There is an issue I suppose of putting something that really shouldn't sit at room temperature all day in there first thing in the morning, although someone above alluded to some really fancy ovens that have built in refrigerators as well. That's above my pay grade. There's also a related "Sabbath mode" on most of these appliances.
That's fair enough, I can see that being useful, especially for those who have jobs that semi-routinely don't hold to any sort of fixed hours.
There can be a happy medium between access and security. I don't think anyone is really arguing the security of checking the status of a device - it's pretty easy to implement that read-only. I can quickly come up with the thought that the network connection could also allow you to activate a function or program but not make changes to it - that would require physical access to the device's control panel.
In your example, you could connect when leaving work and activate the "turn off refrigeration and heat up my dinner at 350 degrees for an hour, then keep warm" function. Maybe at the same time tell your thermostat to change from its "away" to "at home" setting. Someone could break in and do the same, but while you could pay a bit more on your heating bill (because it ran at the higher "home" setting all day) they couldn't make it a LOT more by turning the heat up to 95 degrees.
To each his (or her, in your case) own. Tabs "on top" drive me nuts. But I thought those were made the default some number of iterations ago? I seem to recall spending some time in about:config finding the right tab related variable to put them back pointing down. Personally, I just hope that this remains an option in FF29 and don't 100% force all the UI changes on all of us. I'm not switching to the beta stream to find out.
BTW, I totally agree w.r.t. flat icons, but they seem to be all the rage at the moment.
I can't agree with you that that everyone with over $1million is super-rich.
Here's a very middle-class example: A 50 year old man with three kids under 18. He has $200k equity in his home, and has saved $800k towards retirement. He dies, suddenly. Is it the moral obligation of the government to force the sale of his home and liquidate his retirement account so that they can "wipe the slate clean" for his minor children?
I'll somewhat agree with you that everyone with $1M isn't necessarily super rich, although that puts someone well into the 1% category. Sure, if you live in Manhattan, or SF, or whatnot it's really not an excessive amount of money.
However, the way you wrote what you did suggests that you think that this person's estate will have to liquid on death to pay taxes or make sure his kids have to start from the same place as everyone else. The first $5M or so of an estate (I think, it changes frequently and I'm not in much of a position to worry about it) is free from estate taxes, at least at the federal level. So there's no problem with your subject leaving his entire estate to his 3 kids, largely tax free. That should cover their entire educations with quite a bit left over for a better than good start in real life.
Not only that, but people seem to be forgetting that this is a consumer oriented service, for $1.99/month/TB, and not a corporate oriented service with a real SLA. I also suspect that in general the customer's local network/ISP is going to have issues far more frequently than Google Drive does. For $2/month, I can wait an hour or two to backup or retrieve my stuff, if I need it that critically I'll have a second copy on a thumb drive and/or pay for a better service with a real SLA.
Yes, in this experiment, you'd be hoping that the incentive for the extra cash would outweigh the incentive to do funny business with the federal dollars. You'd have to keep a sharp eye on collusion above all else.
Sadly, given human nature, the more money that's involved is likely to increase the chances of funny business - gotta keep those dollar coming! Not that I'm cynical or anything...
You know, it's not all or nothing. We can and should investigate allegations of fraud/misconduct, but you also bring up some other real problems.
What you call p-hacking is a big problem in more biological/medical fields, I've seen a recent proposal for the grant agencies to fund and require someone be involved who actually understands statistics (and their limitations) with any project that's going to use p-values to "confirm" results.
As for publication bias, I think this takes on 2 forms. The first is that something questionable gets published because the research is question is currently "hot", or because it's related to the editor's field. I'm really not sure how to deal with the first; most journals have multiple editors and I suppose you could recuse one in his/her own field, but at the same time that editor is best able to know who the external reviewers should be. Form #2 is more insidious; a paper gets published pretty much only because the lead investigator is a big name in the field, either solely because "he's big, this has to be good science" or because the editor/reviewer is afraid to recommend major changes or non-publication for fear of retaliation. I've seen this. The best solution, but one I fear most journals will be loathe to implement, is double blind reviewing where each paper is assigned a number when it gets to the editorial office and then the names are removed so it can be reviewed solely for the science. It's not perfect, and in many cases it's possible to have a strong suspicion which lab a paper came from, but would be a huge improvement over what we have now.
There is a push at the moment to replicate results, especially for potentially high impact studies. The problem is, who's going to pay for it? Granted, it should be cheaper to reproduce a result (or fail to do so) than it was to do the original research, but with the NIH payline currently only funding something like 12% of all proposals in the first place, which additional grants are you going to refuse to fund in order to do this? I really don't see the political will to change this, if anything the current budget's slight increase will keep things from getting worse w.r.t. funding. There's also the issue that "original" research is far sexier than reproducing other things.
There's also an issue of WHO is going to do this. I think you'd need to set up an institute that does nothing but replicate results. No PI is going to spend 6-12 months and $10s of thousands doing work that is at most going to get a little note published saying "we got the same result" unless it's critically related to other things going on in the lab. Publish or perish, after all. Also, in niche fields, there may only be a few dozen individuals worldwide capable of doing the work.
Finally, what if the research is on a rare disease that affects 10-20 people/year and requires enough subjects for a clinical trial? Where are you going to find a second set of individuals with the disease, who weren't part of the first trial and are both able and willing to take place in the second?
Really, I'm not arguing that replication shouldn't be done, just trying to point out how impractical it can be.
Yes, biomedical misconduct is against the law, especially when it's funded with federal dollars. ORI only investigates (when bureaucracy allows them to) alleged incidents that are reported to them which only applies to HHS funded research, which is mostly NIH funded research. So it's 100% the business of the federal government. A typical investigation doesn't cost the government that much, anyway, since a lot of it involves making the institution that was awarded the grant (typically a university) conduct most of the investigation and report back, and most will because they want to continue having their other investigators receive NIH funding. Not to mention the university doesn't want to have the reputation of having faculty who misrepresent their research.
Now, if you want to argue that the government shouldn't be funding this research in the first place in which case this office wouldn't be necessary, that's a different story (although I personally disagree with that).
David Malan, who went to Harvard himself and is a rockstar teacher, teaches the course. I watched a couple of his lectures and found them interesting and engaging, even when he covers some basic concepts that I have long known. If I had him teaching me programming back in the day, I might have stuck with it and become a coder myself.
i'm sure its just me, but isn't this possibly the dumbest excuse for not becoming a programmer around?
almost all programmers i know who really add value to projects learned the stuff mostly on their own...teachers don't teach this stuff, the computer does.
Yes, ok, most if not all good programmers learned a lot of what they know mostly on their own, I'll concede that, and I don't think any great coder hasn't learned mostly by doing.
However, a great teacher who can make even mundane topics engaging, can develop a level of interest that makes that future coder actually want to put in the time and effort to learn to code, as opposed to just sitting there watching TV or whatnot.
I don't know about that. I'm far from a power user, and can usually badger GIMP or Inkscape into doing what I need if I'm on a Linux box or at home, but they're nowhere near Photoshop and Illustrator in terms of features or interface. And it pains me to say that, because I'm not a huge Adobe fan and their non-cloud software is way overpriced.
Correct, the (previous, deceased) owner forgot to leave the keys. If this were in meatspace and it was the keys to a door that were lost or forgotten, a locksmith would come in and open the locked door, probably also providing a new key. In this analogy, Apple is the locksmith. As mentioned above, nothing stops them from charging for the service, but they should be able to do this. If it was a locked safe instead of a door, the locksmith might ask for a legal document that the surviving family members are indeed the new owners to prevent liability, and it appears Apple was provided with that or needs to say what additional documentation they need. As for anyone suggesting documents stored on the iPad or in the cloud make this a grey area, how is that any different than a diary, letters, account statements, etc. that might be stored in a physical safe?
I'll vouch for that, my MIL's computer looked almost exactly like that the last time she complained about it being slow. Granted, the teen/tween age nieces had been on their installing lord knows what as well, but there were at least 8 toolbars.
But isn't this (almost) the opposite of what you're talking about? Granted, Dell is making some $ the slimy way by charging to do a quick and easy install of free software that I could probably walk my mom through, but:
All that bloatware you mentioned made money for Dell, etc. because Ebay, Norton, etc. paid Dell to make sure that their software was installed with a prominent icon on the desktop. I don't think this ever increased the cost of a PC to a buyer (ignoring the time and effort to remove said bloatware). In this case, Dell is charging to install Firefox and then pocketing the $ directly, which skates awfully close to charging for FF, and Mozilla sure isn't seeing on pence of it.
It's not just those with compromised immune systems. Quite a few vaccines are produced in eggs, so egg allergy = no vaccination against those diseases. This is particularly relevant for flu vaccines.
OK, so for purposes of writing up a paper and reporting methods, yes, it's not all that free or natural. But in practice it is. You could cut a section from a branch to fit tightly in a tube or funnel, no epoxy needed and no real reason why it would have to be round if it's easier to make a different shape. Wood tends to swell slightly when wet, so it could make itself air tight. Then, while it's convenient to use nitrogen from a tank to provide pressure and simultaneously avoid contamination in a lab setting, you could use a cheap, off the shelf air compressor, or a hand pump/crank system, or press on a plunger, or even passively use the weight from a large enough supply reservoir of water to force water through the filter.
Slashdot Mirror
Somewhat OT, but Thanks for pointing out the bit about the surfactants, which I've suspected is part of the "magic" for a while. Bought some cheap glyphosate weed killer and it didn't work nearly as well as the Roundup, and that's the only difference I could find on the label. FWIW, I think when I added just a little bit of dish detergent to the off brand stuff and gently mixed, it worked a bit better afterwards.
Not only is there "bureaucratic and political interference", but in all likelihood the financing set aside for decommissioning has been "borrowed against for decades. Sort of like surplus funds in Social Security. But we can't possibly make the poor corps operating the reactors actually come up with the money they were supposed to save for the purpose, because that would be bad for business. Nope, I'm not at all cynical...
Glad to hear you find Researchgate useful for your work.
Around here, though, someone set up a Researchgate account and added a couple of publications to their profile. Researchgate then took the "initiative" to send all coauthors invitations to join. Instead of a somewhat spammy message to the effect of "so-and-so has joined Researchgate; we think you might want to check us out", the message was a much more sleazy "so-and-so has invited you to join him/her on Researchgate". Just want the /. crowd to be aware of the sort of tactics Researchgate is using to try to build their member list.
Sounds simple, but for some of us dropping all mail to/from Yahoo groups isn't feasible. Of six mailing lists that I receive messages from semi-regularly for work related purposes, three of them are on Yahoo groups. All are specific to niche software packages and for two of the three at least represent THE primary source of "support" for those packages. Taking your suggestion wouldn't necessarily be career suicide, but could hurt me.
Having said that, I receive those mailing lists at my work address and not through my Yahoo email account, so the DMARC issue isn't likely to affect me too much. I could probably move my personal mail from the latter, but a) inertia and b) I find the newest incarnations of Gmail not enough better than the abomination the Yahoo mail has become to force the issue.
For that matter, the current rectangle would be just fine and dandy if it was keyed. Which would also cost ~$0.
True in general, but not necessarily after you negotiate a large and exclusive deal.
you should be getting rid of any potential vector for badness (any software, particularly software that is known to touch the internet) altogether.
While you have a point, how do you suggest data get moved on and off that old PC? SFTP/SCP touch the internet. USB sticks are a potential attack vector. We can restrict access to only computers on the LAN that lives behind the firewall, but our other computers have to be able to get outside in order for us to do work, which means that our LAN could in theory get compromised.
Seriously, why don't I have mod points when a comment like this comes up. Seriously, +5 Insightful
I work at a university. Around here, lots of investigators have aging but perfectly serviceable equipment of all flavors. Some of it is tied to XP (or Win9x, or DOS) because the software to run the equipment hasn't been updated to run under a newer OS. Some of the equipment is one off or made by a company that's been out of business for years, or there is a newer and fancier instrument so they won't upgrade software for the old stuff and there's no budget to spend >$100k to upgrade a piece of equipment just so it can run newer software.
This isn't limited to Windows, either. The NMR spectrometer I'm chiefly responsible for has parts (including the acquisition computer) that are nearly 16 years old, and the software we use to control it was released in ~2006. That version of the software will only run under RHEL 4.0 with certain specific PC hardware and drivers. When the Dell PC died ~2 years ago, I could either find a similar box online ($350 delivered) OR I could upgrade 1. the computer (cheapest supported workstation was ~$2500) and 2. buy a license for a newer version of the control software (~$10k). That's not the newest software, which only works with newer hardware.
Fortunately, at least our IT and CompSec people are willing to begrudgingly accept the status quo. They are "mandating" that nearly all Windows boxes get update to at least Win 7 by next Monday. However, for those computers that can't, they are willing to accept keeping them offline as much as possible and behind a strong firewall plus using common sense security measures (i.e. not running normal tasks as admin).
Of course there was always the risk that you might be delayed, which of course meant your dinner might burn. I haven't seen this feature in a while...
Any relatively higher end oven has this feature. Mine does - I can set the oven to turn on at the time and temperature I want and then turn off after a set period of time as well. If you're delayed by a lot, maybe it will be cold again when you get home. There is an issue I suppose of putting something that really shouldn't sit at room temperature all day in there first thing in the morning, although someone above alluded to some really fancy ovens that have built in refrigerators as well. That's above my pay grade. There's also a related "Sabbath mode" on most of these appliances.
That's fair enough, I can see that being useful, especially for those who have jobs that semi-routinely don't hold to any sort of fixed hours.
There can be a happy medium between access and security. I don't think anyone is really arguing the security of checking the status of a device - it's pretty easy to implement that read-only. I can quickly come up with the thought that the network connection could also allow you to activate a function or program but not make changes to it - that would require physical access to the device's control panel.
In your example, you could connect when leaving work and activate the "turn off refrigeration and heat up my dinner at 350 degrees for an hour, then keep warm" function. Maybe at the same time tell your thermostat to change from its "away" to "at home" setting. Someone could break in and do the same, but while you could pay a bit more on your heating bill (because it ran at the higher "home" setting all day) they couldn't make it a LOT more by turning the heat up to 95 degrees.
To each his (or her, in your case) own. Tabs "on top" drive me nuts. But I thought those were made the default some number of iterations ago? I seem to recall spending some time in about:config finding the right tab related variable to put them back pointing down. Personally, I just hope that this remains an option in FF29 and don't 100% force all the UI changes on all of us. I'm not switching to the beta stream to find out.
BTW, I totally agree w.r.t. flat icons, but they seem to be all the rage at the moment.
I can't agree with you that that everyone with over $1million is super-rich.
Here's a very middle-class example: A 50 year old man with three kids under 18. He has $200k equity in his home, and has saved $800k towards retirement. He dies, suddenly. Is it the moral obligation of the government to force the sale of his home and liquidate his retirement account so that they can "wipe the slate clean" for his minor children?
I'll somewhat agree with you that everyone with $1M isn't necessarily super rich, although that puts someone well into the 1% category. Sure, if you live in Manhattan, or SF, or whatnot it's really not an excessive amount of money.
However, the way you wrote what you did suggests that you think that this person's estate will have to liquid on death to pay taxes or make sure his kids have to start from the same place as everyone else. The first $5M or so of an estate (I think, it changes frequently and I'm not in much of a position to worry about it) is free from estate taxes, at least at the federal level. So there's no problem with your subject leaving his entire estate to his 3 kids, largely tax free. That should cover their entire educations with quite a bit left over for a better than good start in real life.
Not only that, but people seem to be forgetting that this is a consumer oriented service, for $1.99/month/TB, and not a corporate oriented service with a real SLA. I also suspect that in general the customer's local network/ISP is going to have issues far more frequently than Google Drive does. For $2/month, I can wait an hour or two to backup or retrieve my stuff, if I need it that critically I'll have a second copy on a thumb drive and/or pay for a better service with a real SLA.
any Gen Y and beyond should know how to fiddle with a router.
I think you give the average GenYer WAY too much credit for technical prowess.
Yes, in this experiment, you'd be hoping that the incentive for the extra cash would outweigh the incentive to do funny business with the federal dollars. You'd have to keep a sharp eye on collusion above all else.
Sadly, given human nature, the more money that's involved is likely to increase the chances of funny business - gotta keep those dollar coming! Not that I'm cynical or anything...
You know, it's not all or nothing. We can and should investigate allegations of fraud/misconduct, but you also bring up some other real problems.
What you call p-hacking is a big problem in more biological/medical fields, I've seen a recent proposal for the grant agencies to fund and require someone be involved who actually understands statistics (and their limitations) with any project that's going to use p-values to "confirm" results.
As for publication bias, I think this takes on 2 forms. The first is that something questionable gets published because the research is question is currently "hot", or because it's related to the editor's field. I'm really not sure how to deal with the first; most journals have multiple editors and I suppose you could recuse one in his/her own field, but at the same time that editor is best able to know who the external reviewers should be. Form #2 is more insidious; a paper gets published pretty much only because the lead investigator is a big name in the field, either solely because "he's big, this has to be good science" or because the editor/reviewer is afraid to recommend major changes or non-publication for fear of retaliation. I've seen this. The best solution, but one I fear most journals will be loathe to implement, is double blind reviewing where each paper is assigned a number when it gets to the editorial office and then the names are removed so it can be reviewed solely for the science. It's not perfect, and in many cases it's possible to have a strong suspicion which lab a paper came from, but would be a huge improvement over what we have now.
There is a push at the moment to replicate results, especially for potentially high impact studies. The problem is, who's going to pay for it? Granted, it should be cheaper to reproduce a result (or fail to do so) than it was to do the original research, but with the NIH payline currently only funding something like 12% of all proposals in the first place, which additional grants are you going to refuse to fund in order to do this? I really don't see the political will to change this, if anything the current budget's slight increase will keep things from getting worse w.r.t. funding. There's also the issue that "original" research is far sexier than reproducing other things.
There's also an issue of WHO is going to do this. I think you'd need to set up an institute that does nothing but replicate results. No PI is going to spend 6-12 months and $10s of thousands doing work that is at most going to get a little note published saying "we got the same result" unless it's critically related to other things going on in the lab. Publish or perish, after all. Also, in niche fields, there may only be a few dozen individuals worldwide capable of doing the work.
Finally, what if the research is on a rare disease that affects 10-20 people/year and requires enough subjects for a clinical trial? Where are you going to find a second set of individuals with the disease, who weren't part of the first trial and are both able and willing to take place in the second?
Really, I'm not arguing that replication shouldn't be done, just trying to point out how impractical it can be.
Yes, biomedical misconduct is against the law, especially when it's funded with federal dollars. ORI only investigates (when bureaucracy allows them to) alleged incidents that are reported to them which only applies to HHS funded research, which is mostly NIH funded research. So it's 100% the business of the federal government. A typical investigation doesn't cost the government that much, anyway, since a lot of it involves making the institution that was awarded the grant (typically a university) conduct most of the investigation and report back, and most will because they want to continue having their other investigators receive NIH funding. Not to mention the university doesn't want to have the reputation of having faculty who misrepresent their research.
Now, if you want to argue that the government shouldn't be funding this research in the first place in which case this office wouldn't be necessary, that's a different story (although I personally disagree with that).
David Malan, who went to Harvard himself and is a rockstar teacher, teaches the course. I watched a couple of his lectures and found them interesting and engaging, even when he covers some basic concepts that I have long known. If I had him teaching me programming back in the day, I might have stuck with it and become a coder myself.
i'm sure its just me, but isn't this possibly the dumbest excuse for not becoming a programmer around?
almost all programmers i know who really add value to projects learned the stuff mostly on their own...teachers don't teach this stuff, the computer does.
Yes, ok, most if not all good programmers learned a lot of what they know mostly on their own, I'll concede that, and I don't think any great coder hasn't learned mostly by doing.
However, a great teacher who can make even mundane topics engaging, can develop a level of interest that makes that future coder actually want to put in the time and effort to learn to code, as opposed to just sitting there watching TV or whatnot.
I don't know about that. I'm far from a power user, and can usually badger GIMP or Inkscape into doing what I need if I'm on a Linux box or at home, but they're nowhere near Photoshop and Illustrator in terms of features or interface. And it pains me to say that, because I'm not a huge Adobe fan and their non-cloud software is way overpriced.
Correct, the (previous, deceased) owner forgot to leave the keys. If this were in meatspace and it was the keys to a door that were lost or forgotten, a locksmith would come in and open the locked door, probably also providing a new key. In this analogy, Apple is the locksmith. As mentioned above, nothing stops them from charging for the service, but they should be able to do this. If it was a locked safe instead of a door, the locksmith might ask for a legal document that the surviving family members are indeed the new owners to prevent liability, and it appears Apple was provided with that or needs to say what additional documentation they need. As for anyone suggesting documents stored on the iPad or in the cloud make this a grey area, how is that any different than a diary, letters, account statements, etc. that might be stored in a physical safe?
I'll vouch for that, my MIL's computer looked almost exactly like that the last time she complained about it being slow. Granted, the teen/tween age nieces had been on their installing lord knows what as well, but there were at least 8 toolbars.
But isn't this (almost) the opposite of what you're talking about? Granted, Dell is making some $ the slimy way by charging to do a quick and easy install of free software that I could probably walk my mom through, but:
All that bloatware you mentioned made money for Dell, etc. because Ebay, Norton, etc. paid Dell to make sure that their software was installed with a prominent icon on the desktop. I don't think this ever increased the cost of a PC to a buyer (ignoring the time and effort to remove said bloatware). In this case, Dell is charging to install Firefox and then pocketing the $ directly, which skates awfully close to charging for FF, and Mozilla sure isn't seeing on pence of it.
It's not just those with compromised immune systems. Quite a few vaccines are produced in eggs, so egg allergy = no vaccination against those diseases. This is particularly relevant for flu vaccines.
OK, so for purposes of writing up a paper and reporting methods, yes, it's not all that free or natural. But in practice it is. You could cut a section from a branch to fit tightly in a tube or funnel, no epoxy needed and no real reason why it would have to be round if it's easier to make a different shape. Wood tends to swell slightly when wet, so it could make itself air tight. Then, while it's convenient to use nitrogen from a tank to provide pressure and simultaneously avoid contamination in a lab setting, you could use a cheap, off the shelf air compressor, or a hand pump/crank system, or press on a plunger, or even passively use the weight from a large enough supply reservoir of water to force water through the filter.