This article is exceptionally irresponsible in almost every respect, not to mention laughable in its implicit support for SCO's current Linux IP lawsuits, which are widely seen in the industry as a joke.
I certainly condone neither the nature nor the specific case of the 'MyDoom' attack - it is completely unacceptable. The huge majority of Linux developers would, I am certain, feel the same way.
There is in fact no need for Linux "devotees" to launch a virus-based attack against SCO; in open court, Linux would easily defend the current IP claims brought by SCO, which explains the delay and dithering on the part of SCO in presenting evidence to support their case.
The author even admits "...There's no proof, of course...", so what basis is there for this story at all? Rumour? FUD? It seems remarkably familiar somehow...
Phrases like "It's hard to see how any website could withstand that kind of clever evil", casting the issue in quasi-religious terminology, further undermine the credibility of the author.
In my previous job I worked for a company here in the UK making MRI control systems. The most important customers were Chinese hospitals and universities.
The control software ran on Win98/WinNT, and allowed control of the Imaging System from Excel via ActiveX/COM - no joke!
I inherited development of the application, and it was frightening to see how little the previous developers understood basic programming fundamentals, never mind Windows API or COM: threading was totally screwed up; Delay() calls peppered the code; the COM object was pretty badly misconfigured; and error handling flaky (with try..catch blocks often used just to squelch errors).
To be fair, the hardware had multiple failover/watchdog features, but I certainly would not have put my head inside a scanner controlled by that software!
Precisely. HTTP is simply a convenient protocol to use, because firewalls almost certainly allow it through. Single-click generation of RPC-style "WebServices" apps looks impressive at conferences, but perhaps just saves the developer a bunch of keystrokes.
At DevWeek London *February 2002* Don Box spoke of this RPC-vs-Document conflict, coming out (IIRC) in favour of the Document ("messaging") model.
I'm surprised if the issue really not been largely resolved since then.
Re:Too dismissive of Object Pascal
on
Qt vs MFC
·
· Score: 1
The lack of a single base class in C++ (cf. TObject) is a serious flaw in a language that is supposed to be OO; it forces C++ to jump through flaming hoops (think templates, bizarre typecasting, etc.) to simulate OO.
I'm not claiming that Object Pasal is fully OO - it clearly isn't - but it actually has an object model whereas C++ lacks one. Much of the tantric C++ guru complexity is obviated in Object Pascal by simply casting as TObject.
Too dismissive of Object Pascal
on
Qt vs MFC
·
· Score: 1
So you'd rather have a 'grown-up' but non-OO and kinda ugly language than one with a flexible and powerful object model, RTTI, inheritable message-handling framework, copy-on-write strings, etc., etc.?
All the current hype over.NET is old hat: Delphi has had most of what.NET has to offer for years (not surprising given who wrote them both ).
Object Pascal makes me much more productive than C++ ever did, and I'm not talking about GUI gubbins here, but the 'invisible' back-end stuff.
Nicotine helps prevent Alzheimer's Disease, or at least reduce its severity. My Masters thesis was on this subject, and it seems there is a long-established correllation between smoking and lower-than-expected incidences of A.D. in the general population.
Put simply, nicotine may help because the brain cells destroyed by Alzheimer peptides respond to nicotine-like neurotransmitter; a surplus of neurotransmitter could mean that the remaining brain cells function at a higher rate, thereby offsetting the negative effect of cell death.
The socio-economic arguments against a direct effect of caffeine do not really apply to nicotine - smoking is not AFAIK limited to a particular stratum of society, quite the opposite.
What a choice: "Sanity with lung cancer, or dementia, sir?".
The relevant portion of California Education Code 52720 reads:
In every public elementary school each day... there shall be conducted appropriate patriotic exercises. The giving of the Pledge of Allegiance to the Flag of the United States of America shall satisfy the requirements of this section.
Just about every post in this thread is concerned with details of religious freedom. Doesn't the word patriotic worry anyone in the U.S.? It surprised me that such political brainwashing is actually prescribed by the State; but it shocked me that practically no-one sees this as a Bad Thing.
Our company (Engineering sector, ~30 employees) is about to begin a switch away from M$, and it will happen as follows:
1] Replace all copies of MS Office with [Star|Open]Office, except about 4 users who need advanced features of Excel etc.
Gain: ~$250 per user.
2] Add Linux servers for e-mail, web, SQL, CRM, Process Control, etc.
Gain: no costly M$ licenses. Minimal downtime.
3] Replace the Win2k Servers with Linux boxen running SAMBA. In a company this size, who needs Active Directory? (It's in SAMBA 3 anyhow).
Gain: no further CALs needed.
4] Give anyone who wants one a Linux box to play with, and convert them.
Gain: Louder Linux voice;o)
At this point (say, 1 year from now) we will review the case. The most important savings are in the Office licences and the CALs, which amount to about $300 per user.
The methodology here is one of 'creep': gradually replacing Windows with Linux on all business-critical machines (emphasising reliability, security and cost) will be much easier than a desktop rollout. By the time we might consider Linux on the desktop:
the various Linux desktop environments will be much more refined
we will have plenty of experience with Linux in general
I do not intend to sweat over a Linux rollout, given that the goal is presumably to reduce headaches;o)
There is no such thing as a 1/2 of a firing, which means that there are only 2 states: fire or no fire. This is the definition of a binary system.
I couldn't agree more. My point was that some neurons (esp. in the hippocampus - basket cells???) only fire in response to a repetitive stimulus, not simply to a single 'fire'.
Although the output of a given neuron may be described as binary (as in On-Off), the operation of a network of neurons clearly uses more than the instantaneous values of the individual neurons to achieve State, unlike a conventional binary computer.
His attempts to become a cyborg from what I understand consisted of little more than putting a chip in his body which would open a door as he walked towards it. How is this that different from: having the chip in your pocket, sticking it to your arm with some sort of patch, etc.
I worked in the lab which built the door-opening, PC-booting stuff at Reading. What we didn't tell the countless media hacks was that he had the implant removed after a few weeks, and that a Smart Card in his back pocket was exactly what was opening the doors;o)
He was/is, however, very competent at teaching Control Theory. He had me understanding Nyquist in a few weeks, which is saying something. Unfortunately, as I graduated, he seemed to have laid claim to work done by other people in the department, causing several good staff to leave.
It is a pity that Cybernetics is reduced by Warwick to robotic gizmos, when it should really be known as a meta-science or scientific philosophy. Its applicability is far beyond robots and just the technological, to business models, large-scale human behaviour, meteorology, etc., etc.
The human nervous system is based completely on a binary system.
Although this is the classic textbook view, there is a large body of research to suggest that neuronal firing patterns (frequency, phase, etc.) are important for encoding of information in the brain. For example, many neurons, especially in the hippocampus (~site of memory formation) use a variable number of spikes to encode information, which cannot really be charactarised as 'binary'.
on a second thought; do you have "upstream" nerve channels (hand to brain), and "downstream" (brain to foot) nerve channels? or do they just use the same neural pathways?
'Downstream' nerve signals (*from* the brain) follow what's called the Efferent pathway, and 'upstream' signals follow the seperate Afferent pathway.
Yes. Neurons, in general only carry signals in one direction
Interestingly, it seems that transmission of neuronal potentials in the opposite direction to the main signal is important for learning. This backpropagation effect has nothing to do with neural network learning schemes you learnt in CompSci courses, however.
Nuclear waste storage is very good. It's not like they are hauling it around in thin metal barrels like the environmentalists want you to think. No.
This rather misses the point (in addition to being a bit optimistic). A brief glance at Greenpeace highlights the dangers in long-distance radioactive fuel transport. Trafficking and sabotage of nuclear fuel shipments are the potential source of major disasters, alongside abysmal safety records for fuel storage and reprocessing.
Nuclear power has too many 'collateral' problems, not least in the way it helps the proliferation of nuclear weapons. It's time to ditch it.
Slashdot Mirror
My complaint:
Re: "Linux cyber-battle turns nasty"
This article is exceptionally irresponsible in almost every respect, not to mention laughable in its implicit support for SCO's current Linux IP lawsuits, which are widely seen in the industry as a joke.
I certainly condone neither the nature nor the specific case of the 'MyDoom' attack - it is completely unacceptable. The huge majority of Linux developers would, I am certain, feel the same way.
There is in fact no need for Linux "devotees" to launch a virus-based attack against SCO; in open court, Linux would easily defend the current IP claims brought by SCO, which explains the delay and dithering on the part of SCO in presenting evidence to support their case.
The author even admits "...There's no proof, of course...", so what basis is there for this story at all? Rumour? FUD? It seems remarkably familiar somehow...
Phrases like "It's hard to see how any website could withstand that kind of clever evil", casting the issue in quasi-religious terminology, further undermine the credibility of the author.
In my previous job I worked for a company here in the UK making MRI control systems. The most important customers were Chinese hospitals and universities.
The control software ran on Win98/WinNT, and allowed control of the Imaging System from Excel via ActiveX/COM - no joke!
I inherited development of the application, and it was frightening to see how little the previous developers understood basic programming fundamentals, never mind Windows API or COM: threading was totally screwed up; Delay() calls peppered the code; the COM object was pretty badly misconfigured; and error handling flaky (with try..catch blocks often used just to squelch errors).
To be fair, the hardware had multiple failover/watchdog features, but I certainly would not have put my head inside a scanner controlled by that software!
Nothing is further from the truth. Mailing Disks is Faster than Uploading Data From the linked article: it will take a year to read a 20-terabyte disk
Filesystems most certainly need developing!Precisely. HTTP is simply a convenient protocol to use, because firewalls almost certainly allow it through. Single-click generation of RPC-style "WebServices" apps looks impressive at conferences, but perhaps just saves the developer a bunch of keystrokes.
At DevWeek London *February 2002* Don Box spoke of this RPC-vs-Document conflict, coming out (IIRC) in favour of the Document ("messaging") model.
I'm surprised if the issue really not been largely resolved since then.
The lack of a single base class in C++ (cf. TObject) is a serious flaw in a language that is supposed to be OO; it forces C++ to jump through flaming hoops (think templates, bizarre typecasting, etc.) to simulate OO.
I'm not claiming that Object Pasal is fully OO - it clearly isn't - but it actually has an object model whereas C++ lacks one. Much of the tantric C++ guru complexity is obviated in Object Pascal by simply casting as TObject.
So you'd rather have a 'grown-up' but non-OO and kinda ugly language than one with a flexible and powerful object model, RTTI, inheritable message-handling framework, copy-on-write strings, etc., etc.?
All the current hype over .NET is old hat: Delphi has had most of what .NET has to offer for years (not surprising given who wrote them both ).
Object Pascal makes me much more productive than C++ ever did, and I'm not talking about GUI gubbins here, but the 'invisible' back-end stuff.
Nicotine helps prevent Alzheimer's Disease, or at least reduce its severity. My Masters thesis was on this subject, and it seems there is a long-established correllation between smoking and lower-than-expected incidences of A.D. in the general population.
Put simply, nicotine may help because the brain cells destroyed by Alzheimer peptides respond to nicotine-like neurotransmitter; a surplus of neurotransmitter could mean that the remaining brain cells function at a higher rate, thereby offsetting the negative effect of cell death.
The socio-economic arguments against a direct effect of caffeine do not really apply to nicotine - smoking is not AFAIK limited to a particular stratum of society, quite the opposite.
What a choice: "Sanity with lung cancer, or dementia, sir?" .
Where are the voices advocating political freedom?
from page 3 of the ruling:
The relevant portion of California Education Code 52720 reads: ... there shall be conducted appropriate patriotic exercises. The giving of the Pledge of Allegiance to the Flag of the United States of America shall satisfy the requirements of this section.
In every public elementary school each day
Just about every post in this thread is concerned with details of religious freedom. Doesn't the word patriotic worry anyone in the U.S.? It surprised me that such political brainwashing is actually prescribed by the State; but it shocked me that practically no-one sees this as a Bad Thing.
Don't you guys even suspect that something may be amiss?
Our company (Engineering sector, ~30 employees) is about to begin a switch away from M$, and it will happen as follows:
1] Replace all copies of MS Office with [Star|Open]Office, except about 4 users who need advanced features of Excel etc. Gain: ~$250 per user.
2] Add Linux servers for e-mail, web, SQL, CRM, Process Control, etc. Gain: no costly M$ licenses. Minimal downtime.
3] Replace the Win2k Servers with Linux boxen running SAMBA. In a company this size, who needs Active Directory? (It's in SAMBA 3 anyhow). Gain: no further CALs needed.
4] Give anyone who wants one a Linux box to play with, and convert them. Gain: Louder Linux voice ;o)
At this point (say, 1 year from now) we will review the case. The most important savings are in the Office licences and the CALs, which amount to about $300 per user.
The methodology here is one of 'creep': gradually replacing Windows with Linux on all business-critical machines (emphasising reliability, security and cost) will be much easier than a desktop rollout. By the time we might consider Linux on the desktop:
- the various Linux desktop environments will be much more refined
- we will have plenty of experience with Linux in general
I do not intend to sweat over a Linux rollout, given that the goal is presumably to reduce headachesnullstr
I couldn't agree more. My point was that some neurons (esp. in the hippocampus - basket cells???) only fire in response to a repetitive stimulus, not simply to a single 'fire'.
Although the output of a given neuron may be described as binary (as in On-Off), the operation of a network of neurons clearly uses more than the instantaneous values of the individual neurons to achieve State, unlike a conventional binary computer.
I worked in the lab which built the door-opening, PC-booting stuff at Reading. What we didn't tell the countless media hacks was that he had the implant removed after a few weeks, and that a Smart Card in his back pocket was exactly what was opening the doors ;o)
He was/is, however, very competent at teaching Control Theory. He had me understanding Nyquist in a few weeks, which is saying something. Unfortunately, as I graduated, he seemed to have laid claim to work done by other people in the department, causing several good staff to leave.
It is a pity that Cybernetics is reduced by Warwick to robotic gizmos, when it should really be known as a meta-science or scientific philosophy. Its applicability is far beyond robots and just the technological, to business models, large-scale human behaviour, meteorology, etc., etc.
This rather misses the point (in addition to being a bit optimistic). A brief glance at Greenpeace highlights the dangers in long-distance radioactive fuel transport. Trafficking and sabotage of nuclear fuel shipments are the potential source of major disasters, alongside abysmal safety records for fuel storage and reprocessing.
Nuclear power has too many 'collateral' problems, not least in the way it helps the proliferation of nuclear weapons. It's time to ditch it.