Why is there no verification or personal audit trail available for elections? So that you cannot be held personally responsible by a repressive regime when they find out who you voted for.
Contrary to your suggestion there are techniques that would allow every voter to verify that their vote was counted - yet not disclose the candidate against which the vote was registered. If there was a dummy candidate "abstain" then a well designed system would make coercion at least as difficult as it is today - yet would still improve confidence in the voting system by making it less likely that counting errors could sway the result yet remain undetectable.
I think you completely miss the critical point about the difference between open and closed source in security critical environments. The problem with open source is that everyone has access to the source. The consequence of this openness could easily be catastrophic in the event of a failure.
Any party involved with the deployment of the open-source option is open to accusations of negligence. If they had purchased a closed system they have an obvious defence that it is the supplier at fault.
If the problem is a consequence of an obvious bug then those responsible for technical review and QA could be held liable... This scenario has particularly dire PR consequences if a vocal opponent can demonstrate negligence.
If you buy closed source systems they may have far more serious problems - but you are unlikely to find yourself in political hot water.
I can see both sides of the argument about the parallels between the tobacco industry and the case for legalising drugs. I think it would be a very bad move to commercialize the supply chain... but I also think that prohibition is an exercise in cretinism. For both cigarettes and currently illegal drugs I'd like to see them available (at low cost) from a pharmacy - controlled by similar rules to the supply of poisons. I see no reason why an individual should be prevented from administering any substance to themselves - and only have a real problem with the idea of 'spiking' someone's drink or food without their knowledge. Conversely, I see a need for strict regulation of the supply chain... and a complete ban on the promotion or advertisement of all drugs.
Your argument for the benevolence of drugs companies is flawed too... I'm not suggesting that all the people working for drugs companies are inherently evil but rather that the free market economy encourages practises which don't best benefit society as a whole. The problem, in my opinion, stems from the way in which intellectual property is used to generate revenue. Medical research is massively expensive - and the costs must be met up-front with the expectation of substantial long term profits - which necessarily creates a conflict of interests. In order to appease shareholders, and in order to generate revenue to fund current research, it is inevitable that some will be unable to afford access to medicines. I have no solution to this problem - but I think it is better to be aware that it exists.
Hmmm - interesting - but not what I'm looking for. Ignoring the problem that they are in the states and I'm in the UK, their range is not what I'm looking for.
Their laptops seem to come with Duron/Celeron/P4 processors - where as I'd prefer an (allegedly P4 beating) P4m or centrino based system.
There are no LCDs comparable with my (nearly 2-year-old) UXGA-PLUS 1600x1200 TFT - they have lower resolution screens at the normal ratio or widescreen - which doesn't suit my purposes.
Where wi-fi is included it seem to be 802.11b - whereas I've already got 802.11g - I wouldn't want to take a step backwards.
They would do well to start with developer awareness...I was bemused that while I could find lots of phones touting "support for java games" - I couldn't find satisfactory documentation on what this means. I've no interest in games per-se, however given a blue-tooth phone with support for Java - I am interested in business applications. I want to find out if I can use blue-tooth phones in cryptographic authentication systems; I want to know what APIs are available to allow my phone to run bespoke messaging software. Hell - I'd like to see a sample application which amounts to more than a trivial waste of time. I can't help thinking that this technology holds the key to interesting new systems... but that won't happen if Java is just a buzzword denoting a more expensive toy.
I've an Inspiron 8200 too - similar spec... though I treat mine with kid-gloves. I've had a hard-disk die after about a year which I replaced myself because I couldn't wait for return-to-base or for an engineer to visit - even if I had paid for that service. Apart from this glitch, I'm very happy with my I-8200... 2 years on and it's still credibly quick, and I'm get to find any laptop or desktop with a screen I prefer. The gripe I had when I bought the 8200 was that the Dell upgrades are comically overpriced... but it is fairly straightforward to find 3rd party suppliers for memory / large disks. I was also miffed to find that Dell wouldn't sell me a wireless card **after promising they would when selling the laptop** as Dell seems to only supply mini-PCI cards to the US but not the UK. Oh, and their online support is pathetic when it comes to documenting how to connect SVGA to a TV... the drivers they supply don't match their instructions and (for some as yet unknown reason) I can't get my Inspiron to display on a TV.
While overall happy with the 8200, I seriously doubt I will replace it with another Dell when the time comes. The current Inspiron laptops all come with either lower resolution or widescreen displays. While widescreen displays sound a good idea in practice they are more trouble than they are worth. I love the 8200's 1600x1200 because it allows me to display a lot of text using the standard fonts developed when 640x480 was the norm... widescreen deforms these fonts rendering them unreadable...while widescreen offers more pixels, each pixel is far less useful.
Dell supply cutting edge laptops and frequently for reasonable prices... IMHO, however, their designs today are inferior; their online support is hopeless; they inflate upgrade prices and have utter contempt for their customers.
You're might be tired - but I remain somewhere between amused and incredulous. For example, the BBC recently reported on a campaign to clamp down on music/movie piracy in the UK. Instead of citing lost tax revenues and pointing out the illegality of trading in illegal copies... they play the terrorism card. Apparently Brits should avoid pirate copies because the revenues go to the 'unpleasant criminal underbelly' and funds drug and human trafficking as well as terrorism. Forgive my ignorance, but I was previously lead to believe that the criminal underworld engaged in drugs/human trafficking because those are profitable enterprises in their own right...not a lavish expense supported by MP3/DivX shenanigans!
My anti-spam hiccupped a couple of weeks ago and I saw one of their mails while sipping my first morning coffee. Clearly inflation was a serious problem in Nigeria... they wanted my help shifting 9 billion dollars someone unexpectedly misplaced. How sorry I felt for him trying to launder a sum in excess of the national GDP!
Then you can easily add the features and extensions that you want. AdBlock isn't a required feature for the browser to work so it rightfully shouldn't be included.
Even among the extensions alongside Google-Bar et al? Hmmm.
Personally, I don't see the moral dilemma. Sites are free to reference banner adverts and may choose to do so for revenue. I see no problem with this per se, but I am not interested in the advertising material - so I choose not to display it. If a web site wants to place conditions on my use then it is free to do so, and I would not wish to circumvent those restrictions. I am happy to pay for services I deem worthwhile but prefer not to be bothered with irrelevant nonsense.
I feel the web would improve if everyone blocked banner adverts as this would encourage appropriate advertising. I find it remarkable that advertisers find the blunderbuss approach worthwhile! Good advertisements are content in their own right, and (believe it or not!) I am interested in seeing them, but ONLY when I choose. Web advertisers need to grow up and realise that only a small percentage of the 6-billion potential targets have any interest in their offerings and that everyone benefits when they target their advertising. I look forward to a world in which advertisers look for sites to sponsor where the administration of non-commercial web sites choose to endorse sponsors or otherwise. In an ideal world this would lead to sites presenting only relevant and carefully crafted and seamlessly integrated information and Adblock et al. would become obsolete.
In previous versions of Firefox (when I tried 0.8, for example) Adblock was available as the most recommended extension. Now, after I've advised others try Firefox, and they have installed 0.9, the Adblock extension is no-longer available. I've heard that this was due to compatibility problems with the extension manager - though I've no inside knowledge.
I hope this version either includes Adblock as standard or at least makes it easy to install as an extension. Adblock is a major reason to adopt Firefox - and it was a huge step backwards to find that 0.9 didn't support Adblock by default.
I'm intrigued. I've been well versed in the failures of SQL as an adequate interface to DBMS - and on that score you've been preaching to the choir. However, I remain curious...
Your example seemed very concrete - with your "EXPOSE" syntax you essentially require a new addressing strategy. Detractors suggesting a problem in identifying the type of zip-code you intend to expose miss the point - clearly you intend that the names uniquely identify the domains you intend to expose - for which there are many approaches - even in SQL.
I am concerned by the complexity of an algorithm to establish which relations are to be used to (indirectly) join the exposed domains - I guess only a brute force search for a sequence of relations linking two domains would suffice... this might make it very slow to detect impossible expose queries. Another concern is demonstrated by the following example:
People : PersonID->Name
Friend : PersonId->PersonId
Now consider a named person's friends n-times removed. This simple relational model captures the concept - however it is not straightforward to see how this would be addressed in the EXPOSE syntax.
I was wondering, since I believe the above problems are surmountable, do you have any references regarding a design for EXPOSE - or is this all "off the top of your head?"
I think we can agree to disagree - even if I now disagree by agreeing with your last post.
Writing a good grammar/style checker is nothing short of a devilishly difficult problem - this is probably why we don't have one at the moment. I also agree that it would be foolish for the great Mozilla folks to divert their attention to writing a grammar checker at the expense of their other projects. Maybe the compromise would be to establish a standard API for the integration of 3rd-party style checkers? I'd be impressed if this was done in collaboration with the Open Office folks so any style checker developed would be available to the maximum number of users and attract the widest developer interest.
A few years ago I considered the idea of writing such a tool but abandoned the idea of a solo project for several reasons:
1. I would need to classify words. I don't have the patience to consider the 60,000+ word vocabulary which would be ultimately required to make the project useful. 2. The research required in order to encode patterns/anti-patterns is daunting and it is unlikely that everyone would agree with my subjective decisions. Establishing rules would require some form of consensus. 3. Even if I were to develop a successful tool, it is not clear if this tool could be integrated into the applications I want to use.
While these concerns pose a substantial barrier to just-one-man, it seems to me that a distributed open-source approach may fare better. Given an API for integration into an end-product I'd expect a flurry of pattern/anti-pattern matching engines to be developed to this API, and the emergence of a variety of rule-sets for the engines. I expect at least one successful engine to emerge and for the rule-sets to evolve over time under the influence of feedback from interested parties until - in several years time - we would have a tool which would highlight potentially inappropriate style or grammar with acceptable accuracy.
You picked my least lucid rambling - though I still feel the style was reasonably appropriate for the context in which it was used. If I had an automated tool which alerted me while writing, I may have chosen an alternative.
I'm not convinced that software will be able to make helpful non-trivial suggestions any time soon. Conversely, there is no reason software can't highlight potential problems where the structure of the entered text matches known "anti-patterns". I readily accept your argument that grammar checkers are likely inadequate "when you're trying to express yourself at any kind of advanced level." Conversely, in many circumstances, it is not appropriate to express concepts in an advanced fashion. For many purposes, including the majority of my emails, it is preferable to write in a restrained and boring style as this is more likely to accurately convey the necessary information.
Making positive use of contemporary grammar/style tools is not straightforward -there will always be a conflict between artistic licence and strict adherence to rules. The best compromise I can see is to leave the author in full control but to alert him to constructs which do not conform to a chosen stylistic profile. I believe it is possible to write software which would be extremely effective - until this is available to me I choose the poor substitutes I already have over nothing.
The original topic was email clients. I do think this email requires a new perspective on traditional approaches to written communication. In the past interactions, particularly about technical or otherwise complicated matters, would be extensively verified by both parties before exchange - at least partly because physical documents travel more slowly. Email accelerates the pace and several exchanges are possible each day, hence, formulating replies can be time critical. Under these circumstances mistakes are made even by the most adept. It is important to avoid sending garbled or confusing text - automated proofing can quickly identify potential problems and hence prove beneficial.
Maybe I was unclear that I really want structural proofing tools - not just a grammar checker. I think we agree that the grammar checker in Word stinks - and I understand that you feel the checker is of little benefit yourself. I find the poor grammar checker more useful than none - particularly when there is scant time to review a message before it is sent.
Don't be put-out that I questioned your style. I re-read a few times before I even noticed the oddities of construction - I was not suggesting serious flaws. Conversely, comparing the two sentences:
"How do you think Shakespeare & Dickens and all the other authors pre-20th century managed not to sound like gibbering idiots?"
"How do you think Shakespeare, Dickens and all the other pre-20th century authors avoided sounding like gibbering idiots?"
I prefer the second - I find it clearer. Obviously, this level of attention to detail is problematic when we hope to respond unaided in "real-time" - though, IMHO, this is no reason not to aim for perfection while it doesn't affect deadlines. This is a strong argument for the inclusion of grammar/style verification tools in email suites (and Slashdot!:-)
My own emails rarely resemble the works of Hemmingway or Vonnegut (why you picked this particular author is unclear) - this may be why I find proofing tools extremely valuable? Each to their own, I suppose.
How do you think Shakespeare & Dickens and all the other authors pre-20th century managed not to sound like gibbering idiots? Special typewriter assistants? Magical quills?
Slowly? During a lifetime of painstaking dedication? I suspect famous pre-20th century authors didn't write much email and, as such, remain poor examples in this context. Misuse of tools is more likely responsible for the poor results you have experienced than the concept of automated proofing per se.
Wouldn't you have found it helpful to be shown that "...other authors pre-20th century managed..." might be better expressed: "...other pre-20th century authors managed..."? I'd welcome such a suggestion from future software proofing tools. The inconsistent use of "&" with "and" could be flagged too - not to mention that the former could be safely replaced by a comma. Maybe even a suggestion to replace "...managed not to sound..." with "...avoid sounding..."?:-)
I do not consider a text-book a tool - just one good source of information. Your suggested texts are valid, however, just as software tools are no substitute for education, your text books are no substitute for a quality toolset. Incidentally, as an online reference, I was fairly impressed by the The Economist Style Guide.
I find it strange that you don't distinguish between two distinct scenarios. It is ludicrous to hope an automated checking tool will be an effective alternative to education. It is remarkably small-minded to assume that the educated are no more productive using appropriate automated tools.
I admit that, just as a thesaurus can be misused, so could a grammar checker. I've never argued that these tools alone will improve the quality of prose, but rather that, in the appropriate context, the availability of good quality software tools can improve productivity.
Yes - the Fyshbowl was "my" lab too... until '96 though, I'm afraid I don't remember you by name - just suggesting that you might have looked geeky doesn't sufficiently set you apart from the crowd...
The ADM cursor keys were there to teach you to go-with-the-flow. I always thought of them as an executive gimmick:-)
I agree with BenjyD and babbage that it would be absurd to assume Word could ensure anyone could write grammatically correct documents if they were unable to do so unaided. Conversely, as I had hoped I'd made clear, I don't want it to "help me out with subtle grammatical/stylistic problems" but I still find the (very low quality) grammar checker in Word a useful first-pass proofing tool.
It is obvious to anyone who uses Word extensively that it is trivial to write pathetic drivel that Word thinks is OK - and sometimes (though far less frequently) to write something that is clearly valid about which Word complains. While this severely limits the usefulness of Word's grammar checker to reliably verify grammar, the fact remains that it is of significant net benefit when identifying some forms of obvious structural problem in prose.
Rather than make unhelpful remarks that I should learn to write English properly (which I assume were intended humorous) I would welcome constructive discussion about a better tool. There are dozens of improvements I can think of - here are a few:
Check consistency of spelling (particularly useful for Brits) where, for example, "initialized" and "initialised" are both valid but only one should be used in a related group of documents.
Check consistency of "grammatical person" to avoid awkward association of second-person and third-person sentences.
Cliche identification.
Identification of potential verbiage - for example "policymaking process"; "weather conditions"; "this time around"; "top priority" et. etc.
Identify (inappropriate use of) passive voice - e.g. "a hit b" not "a was hit by b"
While it would be a laudable goal to aim for a tool which would accurately identify correct and incorrect grammar, I agree, this is overly ambitious. Conversely, the grammar checker in Word adequately serves as proof-of-concept that significant benefit can be derived even from a simple system which identifies some structural problems in prose.
Slashdot Mirror
Why is there no verification or personal audit trail available for elections?
So that you cannot be held personally responsible by a repressive regime when they find out who you voted for.
Contrary to your suggestion there are techniques that would allow every voter to verify that their vote was counted - yet not disclose the candidate against which the vote was registered. If there was a dummy candidate "abstain" then a well designed system would make coercion at least as difficult as it is today - yet would still improve confidence in the voting system by making it less likely that counting errors could sway the result yet remain undetectable.
If you buy closed source systems they may have far more serious problems - but you are unlikely to find yourself in political hot water.
Two comments on your comments...
I can see both sides of the argument about the parallels between the tobacco industry and the case for legalising drugs. I think it would be a very bad move to commercialize the supply chain... but I also think that prohibition is an exercise in cretinism. For both cigarettes and currently illegal drugs I'd like to see them available (at low cost) from a pharmacy - controlled by similar rules to the supply of poisons. I see no reason why an individual should be prevented from administering any substance to themselves - and only have a real problem with the idea of 'spiking' someone's drink or food without their knowledge. Conversely, I see a need for strict regulation of the supply chain... and a complete ban on the promotion or advertisement of all drugs.
Your argument for the benevolence of drugs companies is flawed too... I'm not suggesting that all the people working for drugs companies are inherently evil but rather that the free market economy encourages practises which don't best benefit society as a whole. The problem, in my opinion, stems from the way in which intellectual property is used to generate revenue. Medical research is massively expensive - and the costs must be met up-front with the expectation of substantial long term profits - which necessarily creates a conflict of interests. In order to appease shareholders, and in order to generate revenue to fund current research, it is inevitable that some will be unable to afford access to medicines. I have no solution to this problem - but I think it is better to be aware that it exists.
I didn't mean SVGA I meant SVIDEO...
Hmmm... I'd like to write some comprable applications... but I'm coming up against a brick wall finding prerequisite information.
They would do well to start with developer awareness...I was bemused that while I could find lots of phones touting "support for java games" - I couldn't find satisfactory documentation on what this means. I've no interest in games per-se, however given a blue-tooth phone with support for Java - I am interested in business applications. I want to find out if I can use blue-tooth phones in cryptographic authentication systems; I want to know what APIs are available to allow my phone to run bespoke messaging software. Hell - I'd like to see a sample application which amounts to more than a trivial waste of time. I can't help thinking that this technology holds the key to interesting new systems... but that won't happen if Java is just a buzzword denoting a more expensive toy.
I've an Inspiron 8200 too - similar spec... though I treat mine with kid-gloves. I've had a hard-disk die after about a year which I replaced myself because I couldn't wait for return-to-base or for an engineer to visit - even if I had paid for that service. Apart from this glitch, I'm very happy with my I-8200... 2 years on and it's still credibly quick, and I'm get to find any laptop or desktop with a screen I prefer. The gripe I had when I bought the 8200 was that the Dell upgrades are comically overpriced... but it is fairly straightforward to find 3rd party suppliers for memory / large disks. I was also miffed to find that Dell wouldn't sell me a wireless card **after promising they would when selling the laptop** as Dell seems to only supply mini-PCI cards to the US but not the UK. Oh, and their online support is pathetic when it comes to documenting how to connect SVGA to a TV... the drivers they supply don't match their instructions and (for some as yet unknown reason) I can't get my Inspiron to display on a TV.
While overall happy with the 8200, I seriously doubt I will replace it with another Dell when the time comes. The current Inspiron laptops all come with either lower resolution or widescreen displays. While widescreen displays sound a good idea in practice they are more trouble than they are worth. I love the 8200's 1600x1200 because it allows me to display a lot of text using the standard fonts developed when 640x480 was the norm... widescreen deforms these fonts rendering them unreadable...while widescreen offers more pixels, each pixel is far less useful.
Dell supply cutting edge laptops and frequently for reasonable prices... IMHO, however, their designs today are inferior; their online support is hopeless; they inflate upgrade prices and have utter contempt for their customers.
Better talking does not equate smarter!
Tell that to George Bernard Shaw - don't you believe Pygmalion?
"Math is hard." (1990s Barbie)
You're might be tired - but I remain somewhere between amused and incredulous. For example, the BBC recently reported on a campaign to clamp down on music/movie piracy in the UK. Instead of citing lost tax revenues and pointing out the illegality of trading in illegal copies... they play the terrorism card. Apparently Brits should avoid pirate copies because the revenues go to the 'unpleasant criminal underbelly' and funds drug and human trafficking as well as terrorism. Forgive my ignorance, but I was previously lead to believe that the criminal underworld engaged in drugs/human trafficking because those are profitable enterprises in their own right...not a lavish expense supported by MP3/DivX shenanigans!
Your argument only holds if I really was selected to help because I was special. Your argument doesn't account for the other addressees.
(OK - I recalled out-of-date GDP data.)
My anti-spam hiccupped a couple of weeks ago and I saw one of their mails while sipping my first morning coffee. Clearly inflation was a serious problem in Nigeria... they wanted my help shifting 9 billion dollars someone unexpectedly misplaced. How sorry I felt for him trying to launder a sum in excess of the national GDP!
Then you can easily add the features and extensions that you want. AdBlock isn't a required feature for the browser to work so it rightfully shouldn't be included.
Even among the extensions alongside Google-Bar et al? Hmmm.
Personally, I don't see the moral dilemma. Sites are free to reference banner adverts and may choose to do so for revenue. I see no problem with this per se, but I am not interested in the advertising material - so I choose not to display it. If a web site wants to place conditions on my use then it is free to do so, and I would not wish to circumvent those restrictions. I am happy to pay for services I deem worthwhile but prefer not to be bothered with irrelevant nonsense.
I feel the web would improve if everyone blocked banner adverts as this would encourage appropriate advertising. I find it remarkable that advertisers find the blunderbuss approach worthwhile! Good advertisements are content in their own right, and (believe it or not!) I am interested in seeing them, but ONLY when I choose. Web advertisers need to grow up and realise that only a small percentage of the 6-billion potential targets have any interest in their offerings and that everyone benefits when they target their advertising. I look forward to a world in which advertisers look for sites to sponsor where the administration of non-commercial web sites choose to endorse sponsors or otherwise. In an ideal world this would lead to sites presenting only relevant and carefully crafted and seamlessly integrated information and Adblock et al. would become obsolete.
In previous versions of Firefox (when I tried 0.8, for example) Adblock was available as the most recommended extension. Now, after I've advised others try Firefox, and they have installed 0.9, the Adblock extension is no-longer available. I've heard that this was due to compatibility problems with the extension manager - though I've no inside knowledge.
I hope this version either includes Adblock as standard or at least makes it easy to install as an extension. Adblock is a major reason to adopt Firefox - and it was a huge step backwards to find that 0.9 didn't support Adblock by default.
I'm intrigued. I've been well versed in the failures of SQL as an adequate interface to DBMS - and on that score you've been preaching to the choir. However, I remain curious...
Your example seemed very concrete - with your "EXPOSE" syntax you essentially require a new addressing strategy. Detractors suggesting a problem in identifying the type of zip-code you intend to expose miss the point - clearly you intend that the names uniquely identify the domains you intend to expose - for which there are many approaches - even in SQL.
I am concerned by the complexity of an algorithm to establish which relations are to be used to (indirectly) join the exposed domains - I guess only a brute force search for a sequence of relations linking two domains would suffice... this might make it very slow to detect impossible expose queries. Another concern is demonstrated by the following example:
People : PersonID->Name
Friend : PersonId->PersonId
Now consider a named person's friends n-times removed. This simple relational model captures the concept - however it is not straightforward to see how this would be addressed in the EXPOSE syntax.
I was wondering, since I believe the above problems are surmountable, do you have any references regarding a design for EXPOSE - or is this all "off the top of your head?"
I think we can agree to disagree - even if I now disagree by agreeing with your last post.
Writing a good grammar/style checker is nothing short of a devilishly difficult problem - this is probably why we don't have one at the moment. I also agree that it would be foolish for the great Mozilla folks to divert their attention to writing a grammar checker at the expense of their other projects. Maybe the compromise would be to establish a standard API for the integration of 3rd-party style checkers? I'd be impressed if this was done in collaboration with the Open Office folks so any style checker developed would be available to the maximum number of users and attract the widest developer interest.
A few years ago I considered the idea of writing such a tool but abandoned the idea of a solo project for several reasons:
1. I would need to classify words. I don't have the patience to consider the 60,000+ word vocabulary which would be ultimately required to make the project useful.
2. The research required in order to encode patterns/anti-patterns is daunting and it is unlikely that everyone would agree with my subjective decisions. Establishing rules would require some form of consensus.
3. Even if I were to develop a successful tool, it is not clear if this tool could be integrated into the applications I want to use.
While these concerns pose a substantial barrier to just-one-man, it seems to me that a distributed open-source approach may fare better. Given an API for integration into an end-product I'd expect a flurry of pattern/anti-pattern matching engines to be developed to this API, and the emergence of a variety of rule-sets for the engines. I expect at least one successful engine to emerge and for the rule-sets to evolve over time under the influence of feedback from interested parties until - in several years time - we would have a tool which would highlight potentially inappropriate style or grammar with acceptable accuracy.
You picked my least lucid rambling - though I still feel the style was reasonably appropriate for the context in which it was used. If I had an automated tool which alerted me while writing, I may have chosen an alternative.
I'm not convinced that software will be able to make helpful non-trivial suggestions any time soon. Conversely, there is no reason software can't highlight potential problems where the structure of the entered text matches known "anti-patterns". I readily accept your argument that grammar checkers are likely inadequate "when you're trying to express yourself at any kind of advanced level." Conversely, in many circumstances, it is not appropriate to express concepts in an advanced fashion. For many purposes, including the majority of my emails, it is preferable to write in a restrained and boring style as this is more likely to accurately convey the necessary information.
Making positive use of contemporary grammar/style tools is not straightforward -there will always be a conflict between artistic licence and strict adherence to rules. The best compromise I can see is to leave the author in full control but to alert him to constructs which do not conform to a chosen stylistic profile. I believe it is possible to write software which would be extremely effective - until this is available to me I choose the poor substitutes I already have over nothing.
Maybe I was unclear that I really want structural proofing tools - not just a grammar checker. I think we agree that the grammar checker in Word stinks - and I understand that you feel the checker is of little benefit yourself. I find the poor grammar checker more useful than none - particularly when there is scant time to review a message before it is sent.
Don't be put-out that I questioned your style. I re-read a few times before I even noticed the oddities of construction - I was not suggesting serious flaws. Conversely, comparing the two sentences:
I prefer the second - I find it clearer. Obviously, this level of attention to detail is problematic when we hope to respond unaided in "real-time" - though, IMHO, this is no reason not to aim for perfection while it doesn't affect deadlines. This is a strong argument for the inclusion of grammar/style verification tools in email suites (and Slashdot!
My own emails rarely resemble the works of Hemmingway or Vonnegut (why you picked this particular author is unclear) - this may be why I find proofing tools extremely valuable? Each to their own, I suppose.
How do you think Shakespeare & Dickens and all the other authors pre-20th century managed not to sound like gibbering idiots? Special typewriter assistants? Magical quills?
:-)
Slowly? During a lifetime of painstaking dedication? I suspect famous pre-20th century authors didn't write much email and, as such, remain poor examples in this context. Misuse of tools is more likely responsible for the poor results you have experienced than the concept of automated proofing per se.
Wouldn't you have found it helpful to be shown that "...other authors pre-20th century managed..." might be better expressed: "...other pre-20th century authors managed..."? I'd welcome such a suggestion from future software proofing tools. The inconsistent use of "&" with "and" could be flagged too - not to mention that the former could be safely replaced by a comma. Maybe even a suggestion to replace "...managed not to sound..." with "...avoid sounding..."?
I do not consider a text-book a tool - just one good source of information. Your suggested texts are valid, however, just as software tools are no substitute for education, your text books are no substitute for a quality toolset. Incidentally, as an online reference, I was fairly impressed by the The Economist Style Guide.
I find it strange that you don't distinguish between two distinct scenarios. It is ludicrous to hope an automated checking tool will be an effective alternative to education. It is remarkably small-minded to assume that the educated are no more productive using appropriate automated tools.
I admit that, just as a thesaurus can be misused, so could a grammar checker. I've never argued that these tools alone will improve the quality of prose, but rather that, in the appropriate context, the availability of good quality software tools can improve productivity.
Yes - the Fyshbowl was "my" lab too... until '96 though, I'm afraid I don't remember you by name - just suggesting that you might have looked geeky doesn't sufficiently set you apart from the crowd...
The ADM cursor keys were there to teach you to go-with-the-flow. I always thought of them as an executive gimmick:-)
It is obvious to anyone who uses Word extensively that it is trivial to write pathetic drivel that Word thinks is OK - and sometimes (though far less frequently) to write something that is clearly valid about which Word complains. While this severely limits the usefulness of Word's grammar checker to reliably verify grammar, the fact remains that it is of significant net benefit when identifying some forms of obvious structural problem in prose.
Rather than make unhelpful remarks that I should learn to write English properly (which I assume were intended humorous) I would welcome constructive discussion about a better tool. There are dozens of improvements I can think of - here are a few:
While it would be a laudable goal to aim for a tool which would accurately identify correct and incorrect grammar, I agree, this is overly ambitious. Conversely, the grammar checker in Word adequately serves as proof-of-concept that significant benefit can be derived even from a simple system which identifies some structural problems in prose.