This isn't exactly right. When RMS is speaking publicly on behalf of the FSF then of course he is not going to endorse people or products that act in opposition to the stated principles and aims of the FSF. That's no different from any public spokesman: the devil may have the best tunes but you are unlikely to ever hear the pope say "Yes, the devil is a ghastly fellow but there's no harm in dancing with him occasionally, he has such great moves", though of course for Anglicans the situation appears much more nuanced: some dance all night and go back to old nick's "for coffee", some just have a quick shuffle and a grope and worry about being seen, others remain seated but wide eyed and salivating. Old Mark Scuttlebut's users have sore feet ache and coffee breath.

I've heard RMS in interviews say that privately he might recommend Debian to people who want to use a Free Software OS and who appreciate the difference between Free Software and non-free, because he expects they will not enable the non-free sections of the repositories. But of course when speaking publicly as a voice of the FSF he is never going to recommend a distro that offers and perhaps promotes software the FSF exists to make redundant.

Some people will see RMS as a fanatic simply because he does his best to keep to a handful of very simple principles, even if that means inconvenience or ridicule. The interesting thing is that if you wait long enough his fanatical, extremist positions can start to look farsighted and sensible (see GNU/Linux vs Linux naming convention vis-Ã-vis Android, or privacy/data ownership re. software as a service and so on).

I think that makes a good point: appeal to self/other developers being prioritized to the detriment of actual or potential users is a failing that is quite common, and those who do it are naturally enough completely blind to the issue.

Years ago I did use KDE. It could run nicely on the kind of normal hardware most of us have at home, the stuff we buy with our own money and don't throw out after a year. It was also OK on medium and small screens. But KDE4 made it painfully obvious that it was developed by people using very large displays (probably more than one) and very powerful hardware. The huge amount of space taken up by window decoration make it utterly useless on any but the largest laptops (hint: if your photo editor dedicates more display space to (mostly empty!) window decoration than the image then your UI is seriously fucked up), and the gargantuan appetite for system resources means it performs like Vista as well as resembling it. Who actually asked for any of that?

I'm very grateful to the Xfce devs who years ago took a basic but sane UI concept and have stuck to the tasks of incrementally improving it and sometimes integrating new features, while always keeping new versions feeling familiar and never forgetting that the end product has to be something that people like and want and can actually stick with and use every day.

I'm using a SATA connected Hitachi GH15F and cdparanoia. This drive, as far as I can tell, is, or was, an extemely common OEM item.

It works absolutely fine with cdparanoia and, if correct offset is set, gives identical results to EAC in Windows (you need cdparanoia 10.2 or newer; older versions had real deficiencies). I checked this with multiple comparisons where I ripped various CDs, some in poor condition, both with cdparanoia in Debian and with EAC in XP and then md5 hashed the raw pcm output: non-different. I also did rips on different drives on different PCs and achieved bit identical results on those drives which passed cdparanoia -A. Obviously this wasn't a huge dataset and doesn't prove anything but it was good enough for me to stop caring any further.

Here is the output of cdparanoia -A:

CDROM model sensed sensed: HL-DT-ST DVDRAM GH15F EG00

Checking for SCSI emulation...
        Drive is ATAPI (using SG_IO host adaptor emulation)

Checking for MMC style command set...
        Drive is MMC style
        DMA scatter/gather table entries: 167
        table entry size: 524288 bytes
        maximum theoretical transfer: 37074 sectors
        Setting default read size to 27 sectors (63504 bytes).

Verifying CDDA command set...
        Expected command set reads OK.

Attempting to set cdrom to full speed...
        drive returned OK.

=================== Checking drive cache/timing behavior ===================

Seek/read timing:
        [47:10.36]: 55ms seek, 0.36ms/sec read [37.4x]
        [40:00.33]: 61ms seek, 0.39ms/sec read [34.6x]
        [30:00.33]: 51ms seek, 0.42ms/sec read [31.9x]
        [20:00.33]: 51ms seek, 0.48ms/sec read [27.7x]
        [10:00.33]: 63ms seek, 0.58ms/sec read [23.1x]
        [00:00.33]: 66ms seek, 0.74ms/sec read [18.0x]

Analyzing cache behavior...
        Approximate random access cache size: 16 sector(s)
        Drive cache tests as contiguous
        Drive readahead past read cursor: 234 sector(s)
        Cache tail cursor tied to read cursor
        Cache tail granularity: 1 sector(s)
                        Cache read speed: 0.16ms/sector [85x]
                        Access speed after backseek: 0.71ms/sector [18x]
        Backseek flushes the cache as expected

Drive tests OK with Paranoia.

As you can see it isn't going to be quite as fast as your old IDE drive but it isn't exactly slow either.
You can safely ignore fetishists who feel EAC is magically unique and that cdparanoia can't do secure ripping. It can, as long as the drive passes the cdparanoia -A test. If you feel the need to compare your rips with rips made by properly configured EAC or dbpoweramp or similar then you need to set the offset correctly.

Almost all the cdparanoia GUI's ignore the offset and don't allow the user to set it, so their rips will have a different checksum than an offset corrected rip by other tools. This doesn't have any bearing on the quality of the rip, only on the ability to compare it. It hasn't done much for cdparanoia's reputation but if you use it with a fully configurable command line front end such as ripit or abcde, or just by itself, you can get 100% secure rips equally good as those produced by magic tools with proprietary voodoo and vociferous fanboys.

ripit is a perl script front end to cdparanoia, it will:

"do the following without user intervention:

getting the audio CD Album/Artist/Tracks information from MusicBrainz or freeCDDB,
ripping the audio CD Tracks,
encoding to Flac, mp3, Ogg-Vorbis, mpc, m4a or als,
id3 tags encoded songs,
creating an playlist (m3u) file,
optionally generating a toc or inf files for DAO burning with, CD-text
optionally preparing and send a CDDB submission and save it locally,
optionally extracting hidden songs and split ghost songs,
optionally creating md5sum files for all tracks,
running several encoder processes at the same time and same run."

Recent versions will also embed covert art in the tags. Basically, if you trust the CDDB data, you can automate everything and even run it in a loop so all you have to do is swap the CD when the tray pops open and then push it shut again. Get a slot loader and become maximally lazy ;-)

http://www.suwald.com/ripit/news.php
xiph mailing list discussion cdparanoia compared with EAC: http://lists.xiph.org/pipermail/paranoia/2009-June/001574.html

Xfce allows the user to switch off compositing in the settings GUI or, more usefully for scripts and launchers, with a command:

Compositing off: "xfconf-query -c xfwm4 -p /general/use_compositing -t bool -s false"
Compositing on: "xfconf-query -c xfwm4 -p /general/use_compositing -t bool -s true"

A KDE dev pretending that Gnome 3 or Unity are the only other options makes him seem slightly desperate way.

"I'm going to call BS.." isn't informative, nor a convincing refutation.

There is some more info at http://wikibin.org/articles/adam2.html and at http://ar7.wikispaces.com/ADAM2

For example

"ADAM2 maintains a set of ''Environment Variables'' in a so-called Non Volatile RAM, emulated by the last FLASH partition mtd3. This partition is split in two : the first 10kB contain the ADAM2 environment variables, whereas the last 54kB store a XML file detailing the global configuration of Montavista Linux as it was saved by the user."

A memory leak is fixed by rebooting. This isn't about memory leaks, or volatile RAM being fragmented, it's about an file system on "disk" being fragmented.

If you've used different router firmwares, especially the third party ones, you'll have seen that other types of router also suffer some fragmentation but the operating system routinely performs effective automatic defragmentation of the environment. As far as I know ADAM2 is the poorest performer in that it gets fragmented easily, doesn't repair itself automatically and can reach the point where it is beyond repair, all this just by the user upgrading the firmware and changing settings. But that doesn't mean it's the *only* poor performer and it doesn't seem unreasonable to posit that the apparently common phenomenon of wireless transmission devices (soho wireless router) performance degrading over a couple of years might be associated with similar but less severe (or better managed) fragmentation of non volatile RAM.

Flash memory fragmentation is a problem with lots of older routers (and maybe newer ones too, I'm not sure). A fragmented environment can cause all kinds of degradations from poor wireless performance right up to ethernet ports failing and the entire device failing beyond repair (I have experienced this).

There is some good info at http://www.routertech.org/firmware-faq/

RouterTech is a good site for networking info and also offers a FOSS Linux based replacement firmware for Texas Instruments AR7 based ADSL modem/routers.

"Q. I have heard about fragmented flash memory (or environment) in routers. What is this, and how do I deal with it?
A. Flashing firmwares, saving configuration settings, and doing stuff with environment variables all involve writing to the router's flash chip. Over time, the flash memory (particularly the area holding the configuration information and the router's environment variables - i.e., the first 10kb of the router's mtd3 partition) can become fragmented. If this happens, you can have all sorts of problems. The most common ones include not being able to upgrade your firmware successfully via the web interface, not being able to save your configuration settings successfully, and routers bricking themselves spontaneously. This is a major issue on routers with the Adam2 bootloader, which is seriously broken. In our experience, it is particularly problematic with DLink Adam2-based routers - but all Adam2-based routers suffer from this problem, because the bootloader cannot defragment its environment properly, except manually from the Adam2 bootloader command prompt itself. The PSP bootloader, on the other hand, does the job pretty well by itself. "

"They need to get past their upsetness over the video, it's just one angry guy's creation."

Sounds a lot like Islam!

"Both findings gibe with the U.S. Census Bureau's American Community Survey"

Gibe doesn't mean what the author thinks it means, in fact it's not all that far from the polar opposite.

"WordNet (r) 3.0 (2006) (wn)
gibe
        n 1: an aggressive remark directed at a person like a missile
                  and intended to have a telling effect; "his parting shot
                  was `drop dead'"; "she threw shafts of sarcasm"; "she takes
                  a dig at me every chance she gets" [syn: shot, shaft,
                  slam, dig, barb, jibe, gibe]
        v 1: be compatible, similar or consistent; coincide in their
                  characteristics; "The two stories don't agree in many
                  details"; "The handwriting checks with the signature on the
                  check"; "The suspect's fingerprints don't match those on
                  the gun" [syn: match, fit, correspond, check,
                  jibe, gibe, tally, agree] [ant: disaccord,
                  disagree, discord]
        2: laugh at with contempt and derision; "The crowd jeered at the
              speaker" [syn: jeer, scoff, flout, barrack, gibe]"

...when you take it from my soft, warm, vomit stinking, slippery, snotty, infectious podgy hands.

waaaaah.

I am 45 and three quarters.

"However if I have permission to view that email and leak the contents, they've no way of really tracing it back to me, especially if I'm part of a team"

Absolutely not true. I've worked in environments (a large corporation in UK and a regional police force) where data protection is taken seriously. In each case I was part of a team and routinely handled private and confidential information (and of course had signed an acknowledgement of the official secrets act). Every single time I accessed data, even if only to view it, I was and remained identifiable. Everyone was audited both routinely and randomly and any unusual or unexpected access prompted specific inquiries from above. A grossly inappropriate access could result in dismissal. I have seen this for myself after some jealous sap did some sniffing around accounts of their ex partner's new companion - they were escorted from their desk by company security, suspended and subsequently prosecuted, convicted and fired. One colleague accross the desk from me dealt with an account enquiry from an *extremely* prominent person and her jaw nearly hit the desk when 5 minutes later the company security office was on the phone making sure she had been accessing and modifying the account legitimately. These things *can* be done properly and it's ridiculous to imagine that an employee can navigate their way around audited company or government systems without leaving an identifiable trail.

If you communicate via webmail or sms then there are already third parties (the organisations and people who provide those services) who can read your communications, and if the messages are not encrypted then there are several more places along the route where the content can be captured.

It is courting disaster to conduct confidential or government business via domestic webmail, sms or any services that are not accountable, auditable and vetted, and hosted by companies that are subject to UK law.

The fact that people within government departments and political parties gossip/leak, whether inadvertantly or deliberately, doesn't make it a good idea to conduct government business on hotmail or gmail or chat. It means that the existing regulations and laws are not being observed. The sanctions and remedies in those cases are already available and it's the responsibility of officials and politicians to enforce this, *not* to try to get around it by conducting public business in secret.

The fact that the Brown family's intimate private affairs were leaked or stolen makes for a very good illustration. Those thefts came from his *personal* affairs, not from government departments. In another instance the Sunday Times managed to access his *personal* legal and bank files. ref: http://www.bbc.co.uk/news/uk-politics-14119225

From the article:

"New government guidance will say the act relates to "the nature of the information and not the format"

Which means that communications re government business will fall under the Freedom of Information Act, that is they will be discoverable, regardless of the medium of transmission/storage. Personal exchanges will remain private.

This doesn't mean that anyone gets to read ministers' personal/domestic/private sms or emails /partners but it does mean that ministers can't hide official business under unofficial accounts.

Another benefit is that it might discourage ministers/officials conducting business on accounts which are hosted by potentially insecure and definitely unaccountable webmail providers, many of which would be storing data and hosting services outside the UK.

"Windows share? NFS? FTP?"

Making these shared is a deliberate act.

"Vulnerable machine, perhaps not updated properly?"

That would be negligent. If I fail to maintain my vehicle and I crash it into your garden fence because the brakes failed I am still responsible. I can try "I forgot to check the brakes work" as a reason/excuse in the correspondence with the insurance company but it is doomed to failure, along with "the dog ate my homework" and "it was my twin" and "I thought the fence was communal property with me being a part owner".

"We're assuming someone borrowing yer wireless, yes?"

No, we're not. According to the article the acquisition of copyrighted works was done by a member of the household (the account holder's wife) but the account holder was eventually fined for failing to secure his wireless access point. There was no suggestion that any unknown 3rd party used the access point. Please remember that this was a third strike penalty, so notice had been given. The man failed to control his wife, who likes Rihanna. Some kind of penalty is not only inevitable but, if it helps one husband have a quiet life and one apartment block become free of the sound of generic R&B, also a kindness for all concerned and a public service.

"accidentally left a drive open"????

I use p2p applications such as bittorrent (Transmission and rtorrent clients) and ed2k (aMule). Since the demise of Kazaa and similar there hasn't really been a circumstance where the p2p user might have "accidentally left a drive open". If I get caught I will be pissed off but I won't really have a right to complain if the penalty is proportionate.

I didn't make any claims about self representation, so your attempt to take me to task for this is a straw man argument.

The person has been penalised because they are responsible for an access point that has been used to acquire copyrighted works in breach of the copyright holder's rights. The penalty isn't heinous or draconian, it's of the kind that will be slightly uncomfortable and probably memorable. What is so unfair?

The penalties imposed in the USA are vicious, vindictive, disproportionate, greedy, shameful, life destroying and disgraceful. The penalty in France is proportionate and rational: no special interest group is gaining vast unearned riches by collecting it, and nobody's life is being crushed by suffering it. This is exactly how the law should operate.

http://yro.slashdot.org/story/12/09/11/1740241/8th-circuit-upholds-220000-verdict-in-jammie-thomas-case

In the USA it's $9250 per song. In France it's â75 ($190 US) per song.

The penalty in France seems to me to be proportionate and sane. The person penalized did, or allowed to be done, something illegal but not especially malicious or very damaging. They face a penalty which will certainly be unwelcome and which will probably encourage them to act within the law. No huge court case, no lives wrecked, no lawyers riding the gravy train. *This is how a legal system is supposed to be.* That is the difference between "The Rule of Law" and "The Rule of Lawyers".