Although I have a lot of difficulty to connect the reporting about this study which you have linked to with the actual conclusions from the published study. The study rules out a certain mechanism for damage from low-dose radiation which has been hypothesized, but does not say that damage does not exist. Quite the opposite:
"We note that despite the minor direct impact of radiation on redox status of the cell and on antioxidant concentrations, it is well known that even low dose ionizing radiation can cause negative effects via DNA damage. Such damage is direct—caused by strand breaks and deletions—or indirect, from the free-radical products of water radiolysis in the immediate vicinity of nucleotides. At dose rates of order of 417 Gy h1 (representing the most contaminated parts of the Chernobyl exclusion zone), radiation effects on organisms would be expected, and have indeed been observed [16,17]. The present study shows that observed effects are unlikely to be due to radiolysis products directly causing oxidative stress, significantly clarifying discussions about low-level radiation and oxidative stress."

Also note the clear statement "it is well known that even low dose ionizing radiation can cause negative effects via DNA damage". Which is indeed well known, but disputed for example by Mr D.

Anyway, wildlife near Chernobyl certainly benefits a lot from not having humans close by anymore, so probably is better off now than before Chernobyl - despite increased radiation levels.

Nonsense. Both studies provide direct empirical evidence.

Pearson et al., The Lancet:

- "We did a study to directly assess the question of whether cancer risks are increased after CT scans in childhood and young adulthood."
- "In this retrospective cohort study, we show significant associations between the estimated radiation doses provided by CT scans to red bone marrow and brain and subsequent incidence of leukaemia and brain tumours."
- "We noted little evidence of non-linearity of the dose-response, using either linear-quadratic or linear-exponential forms of departure from linearity (leukaemia exponential p=02672 and quadratic p=04683, brain tumour exponential p=09203 and quadratic p=08993)."

Mathews et al., BMJ:

- "In this paper, we derived direct estimates of the increased cancer risk in the first decade or so after CT scan exposure by comparing cancer incidence in over 680000 people exposed to CT scans at ages 0-19 years with cancer incidence in a comparison cohort of over 10 million unexposed persons of similar age."
- "Our results are also generally consistent with the linear no threshold theory (that is, there is no threshold dose below which there is a zero risk)."

First, the first citation is published under "Reviews and Commentary". I added it because I thought it might be easier to understand. This means you are critizing the Lancet study because somebody commenting on it in a seperate publication does not mention "margin of error" or "control groups". The purpose of this commentary is to put it into perspective, not to repeat the data. If you want to critize this study, you should go to the original (it is the second link I gave). You will find all the data there, and also the results are in agreement with the LNT model:

"We noted little evidence of non-linearity of the dose-response, using either linear-quadratic or linear-exponential forms of departure from linearity (leukaemia exponential p=02672 and quadratic p=04683, brain tumour exponential p=09203 and quadratic p=08993)."

And to make this clear, I am not going to discuss the merits of these studies with some random person on the internet. If you find problems with these studies, please write letters to the editors of Lancet and BMJ etc.

Finally, nobody is using "scare tactics" here. I am myself not arguing with you because I think the health impact from nuclear power is serious problem. In my opinion, nuclear power is very safe. I am arguing with you, because you clearly misrepresent what the current scientific consensus is,
based on some questionable information you found on the internet when googling for things which could confirm your existing opinion. This is a dangerously misleading way to build an opinion.

Two of the studies I quoted were not predictions. The cancer from CT has been predicted a *long time* ago. These are the studies *confirming* the predictions using large scale statistical studies.

The study about Fukushima was a prediction based on the LNT model, I cited it because it is relevant in the context of this discussion and to show that the LNT is taken seriously by many different people (including researchers at Stanford - a respected scientific institution).

You may read such nonsense on the internet - but the scientific consensus is actually in favor of the LNT model. Allmost all evidence we have (and this does not only include the Japan bomb research, but for example the studies I cited about CT) is in argreement with the LNT model.

Again, the CT studies are not predictions - but large scale statistical studies which confirm the effect predicted by the LNT model.

Irrelevant. I was correcting Mr. D.'s claims, not commenting on anything else.

"that statistical real world evidence of actual cancer cases attributable to CT scans is non-existent"
I gave you empirical studies published in highly regarded journals which provide exactly such evidence. You counter this with some blog post (which I didn't read) you found on the internet. You don't see what is wrong here?

Since you are obviously cherry-picking your sources again (which I have pointed out to you before), let me add some recent sources from highly respected journals about the risk of low-dose radiation. Ofcourse, according to Mr. D. all these journals just publish pseudo-science. Reminds me of the old joke with the wrong-way driver.

"... First, it is clear that we have now passed a watershed in our field, where it is no longer tenable to claim that CT risks are "too low to be detectable and may be non-existent" (5). A large well-designed epidemiologic study has clearly shown that the individual risks are small but real..."
Journal: Radiology
Link: http://pubs.rsna.org/doi/full/...

"...We noted a positive association between radiation dose from CT scans and leukaemia (...) and brain tumours (...)."
Journal: The Lancet
Link: http://www.sciencedirect.com/s...

"Conclusions The increased incidence of cancer after CT scan exposure in this cohort was mostly due to irradiation. ..."
Journal: British Medical Journal
Link: http://www.bmj.com/content/346...

"The study supports the extrapolation of high-dose rate risk models to protracted exposures at natural background exposure levels."
Journal: Leukemia
Link: http://www.nature.com/leu/jour...

And with respect to Fukushima there were recent estimates from a Stanford guy:
"We estimate an additional 130 (15â"1100) cancer-related mortalities and 180 (24â"1800) cancer-related morbidities incorporating uncertainties associated with the exposureâ"dose and doseâ"response models used in the study. We also discuss the LNT model's uncertainty at low doses. .... Radiation exposure to workers at the plant is projected to result in 2 to 12 morbidities. An additional [similar]600 mortalities have been reported due to non-radiological causes such as mandatory evacuations."
Journal: Energy & Environmental Science
Link: http://pubs.rsc.org/en/content...

Yes, and isn't great that you can do this?

But should Linux drop X many applications will stop supporting X properly. They will then not run properly on any X server anymore, neither on Mac, Windows, or Linux or elsewhere. Or in other words, your X server on Windows or Mac OS X is only useful, because there is currently a large ecosystem based around X.

If Linux switches to Wayland, this ecosystem will be gone. X currently offers compatibility across different architectures, along time (currently, you can still run decades-old X application just fine), and space (network transparency). X as a standard provides as much value as POSIX. Why do you want to break this?

On the few new Linux-based mobile platforms which currently use Wayland, X compatibilty is alread lost. Just sad.

And what do we gain if we replace X? Will it be faster? No, Wayland has basically the same design as X: Message passing using a UNIX domain socket and buffer sharing for direct rendering. Performance wise, there is not really anthing to gain. X is bloated? Do you really think a few kilobytes of old and unused rendering code needed for backwards compatibility are bloat? The design of X is unfixable? Nonsense, X was designed from the beginning on to be extensible. It would be very easy to add a special screensaver extension, if really needed.

Yes, Ubuntu still uses X. Maybe they are using Mir on their phone OS or something?

This straightforward solution is that time travel does not exist.

I don't think so. I am actually much more afraid that we actually get Wayland soon by default, but gradually lose backwards compatibility to rarely used but hard-to-replace applications (and of course network transparency). I seriously do not see that we will gain anything.

Why would you #ifdef out assertions in production code?

Ofcourse... Dream on.

Exactly. That you should only use ssh to tunnel X and only between trusted hosts is well known. It would be nice if you could run untrusted clients on X (and the X security extension was meant for this), but nobody seems to work on this. This would be vastly more useful IMHO than re-building everything on top of a dumbed down protocol: Wayland.

The solution the Wayland guys offer for remote desktop: Use RDP. As if this proprietary protocol from Microsoft never had security problems....

Also, for a different perspective. Look at this:
http://media.ccc.de/browse/con... ... and don't jump to conclusions based on the title. Just watch and pay attention especially with respect to the comments about security of core X11 vs. Qt. And then maybe don't use KD anymore.

In my opinion, breaking compatibility with the X protocol would be the biggest strategic blunder Linux community coud do. Even bigger than messing with the GUI in stupid ways exactly when everybody using Windows is frustrated with the GUI

Without exceptions, you would put in an assertion which would give you the same debuggability (and the compiler would warn if you completely forget to handle the return code). The only advantage of exceptions is that you can ignore the error on the intermedate level and try to handle it higher level instead. This might be useful, but opens up an entire class of new - and very hard to debug - failures from code which is not exception-safe. RAII helps with this, but has its limits.

I understand RAII just fine.

In fact, what you say seems to be in agreement with my statement: "Provided you never have cyclic mutable graphs" "only some memory and resource leaks go away".