Hacker News Reader

Comment by tut-le-tut | Hacker News Reader

"At that rate, the 10 employees would have to work non-stop 24 hours a day, seven days a week to produce the 55,000 pages a month (and would still fall a bit short)."

So 50 people on a normal (40 hour work week) could do this in a month. The FDA has 18 000 employees. I think the answer is clear.

View on HN

If they want their documents faster they should just pay for it. Requesting 300k+ pages is not a normal request, there is no way all of those pages will be of interest.

Pay for it by paying a fee equal to the cost of hiring / training additional personal (https://www.foia.gov/faq.html).

So pay for transparency that is already expected? Should you have to pay to get your medical record?

It's asinine to think you have a right to see every document the government has including PII and trade secrets, even worse to base an anti-vaxx effort on such ideas.

It's crazy to me the amount of people that people would blindly trust the same government that did illegal experimentation in the past and gave unsuspecting populations syphilis. Or the same companies that paid billions for lying in the past.

Only the rich should get FOIA?

The FOIA should not allow anyone to launch a denial of service attack on the government.

The requester can specify the scope or the timeline. Not both. If my employer asked me to review this much content I would also give a timeline on the order of years. Saying "too bad, it has to be faster" doesn't change my estimate.

DOS attacks were pioneered by progressives in the 60s. We’ve had 60 years to make the system robust to DOD attacks

I have no problem with that, but if the paperwork isn’t produced, the FDA can’t enforce or use any of the process for any reason because its not transparent.

Take as long as you want, just lose any legal power those documents give you so that you aren’t unjustly using them if there is evidence of some kind of fraud or other issues.

Take 100 years, but for those 100 years the paperwork can’t be used with any legal authority because its not transparent.

Problem solved.

That is once again absurd and amounts to a denial of service attack. Reviewing documents for public disclosure, removing PII, etc. is a totally different process from reviewing their accuracy for medical purposes. It takes at least as long, probably longer.

Or maybe, if you think the FDA did something wrong, you should have enough of an idea of what it did wrong that you could confirm it by asking for something less than literally everything Pfizer ever submitted to the FDA? If you're asking for that much information, that suggests that you're undertaking a fishing expedition rather than actually substantively accusing the FDA of something wrong.

Yeah, but you're allowed to fish, even if what you're fishing for is dumb. The obligation is for the government to provide transparency, not for it to proctor investigations.

We're spending TRILLIONS of dollars on this pandemic. TRILLIONS! These FOIA requests are not DOS attacks. It's very clear that some are very scared of what these FOIA requests might reveal rather than a concern about the cost of spending 0.000001% of the response budget on it. Why are you personally so scared of what the documents might reveal? It's nothing to do with imaginary DOS attacks.

I am all for the documents being released eventually, but it takes time to remove PII from 450,000 pages of medical records (themselves very dense with PII). I am not in favor of releasing them prematurely with identifying information, nor am I in favor of requiring an entire department to stop their (more important than this process) jobs at the whim of someone who doesn't even know what they're looking for.

If by we you mean the FDA then no they are not. Additionally said trillions aren't allocated to redacting documents so it really is like saying w're spending trillions on the military so deploy the national guard to redact the files. It's apples & oranges.

As outlined in the article, the FDA only has 10 people trained to do redactions and that's for all FOIA requests. Requesting 10 people to redact ~300k documents is going to delay (aka deny service) to anybody else requesting FOIA requests for the next year.

Only the rich should be able to make unreasonable requests (its kind of how Capitalism works).

One trivial way for a normal person to do this FOIA would be to make a request for the document names that lead to the decision to approve Pfizer's vaccine. Redacting a ~10 word title is going to require way less time than redacting ~300k files. We also already know that the FDA knows what files lead to the decision (~300k) and which they are (cause they're redacting them atm). Once you have this list you can decide which files could help your case and request those.

As far as transparency, the FDA didn't just write down two words "Pfizer Approved" they've created document outlining why they approved the vaccine.

https://www.fda.gov/emergency-preparedness-and-response/coro...

From earlier in the article, the total number of requested pages is 450,000. At 8 minutes per page, it's 60k man-hours, or 30 man-years, of work.

That's not how labor works. Adding more people often times increases the time a task takes.

Well, this task is highly parallelizable. Is it not? Why would there be a central bottleneck that you could not scale out?

There are literally legal firms that employ hundreds of paralegals who do nothing but bulk redaction. They literally have systems to split up huge document bundles and do redaction in parallel.

That adage is meant for tasks that require a large amount of coordination, e.g. large software projects. This is no such task.

It's like saying you've got a large wall that needs to be scrubbed clean, and having 10 people do it at once would be slower than having one person do it. That is obviously not going to be the case.

You're roughly talking about the trusts popularized as Amdahl's Law [0]. Which, to summarize the nuances in this context, says "It depends on the nature of the task."

[0] https://en.m.wikipedia.org/wiki/Amdahl's_law

Reply on news.ycombinator.com