Re:VB already gets the respect it deserves...
on
Lisp and Ruby
·
· Score: 1
I think when they say VB is OO they mean it the way people talk about Scheme being OO (that is it supports object design methods, that is data structures can have functions associated with them. I don't think they mean its OO in the sense that it uses the Java/Smalltalk paradigm of syntax. Further the LINQ direction seems to be to move more towards the Lambda family of languages; which given C#'s move in the opposite direction makes sense. Why should Microsoft only support one programming paradigm?
Re:VB already gets the respect it deserves...
on
Lisp and Ruby
·
· Score: 1
Who cares whether its a technology you would have chosen? What you have with the VB program is:
1) A fully working prototype 2) Business rules and logic spelled out and worked through 3) A GUI interface agreed to with the customer 4) Clear limitations of the above model understood. 5) An understand of how this app is going to mature overtime
Why doesn't that justify a rewrite at this point?
Re:VB already gets the respect it deserves...
on
Lisp and Ruby
·
· Score: 1
How much percent of actual professional VB developers is out there?
In 2007 a very low number. In 1995 a much higher number. In 1995 there was for all practical purposes no Java. GUIs were very platform specific and very complex requiring a great deal of understanding of underlying OS concepts. C/C++ required programmers to work as such a low level that business rules were obscured, hat is programmers were something like 15x as productive in VB as they were in the C++ GUI languages.
Corporate American had definitely moved towards the Windows desktop without thinking through a strategy for writing a huge quantity of client programs quickly enough that they wouldn't need a major round of mainframe / mini computer upgrades. In that world VB, Lotus/Excel, Foxpro/Access/Paradox programming comprised something like 70% of all development.
Re:VB already gets the respect it deserves...
on
Lisp and Ruby
·
· Score: 1
te a horrible unmaintenable mess which not even the original author will touch, and which has is almost certainly going to be rewritten by a developer at some point in future
That would actually be a perfect end for a VB program. After the iterative process of requirements gathering is done via a VB rapid development methodology it wouldn't be at all unreasonable to then come in and rewrite it. The rewriters now have:
1) A fully working prototype 2) Business rules and logic spelled out and worked through 3) A GUI interface agreed to with the customer 4) Clear limitations of the above model understood.
So what exactly went wrong?
Re:VB already gets the respect it deserves...
on
Lisp and Ruby
·
· Score: 1
Because that wasn't the intention. VB (VB 3-6 not.NET) apps were meant to be used for short term and limited goals. If a program is going to be used for an extended period of time and modified it shouldn't have been written in VB. Clarity of concept requires organization of concept. VB was designed to work well for an iterative process of discovering the organization, that is VB is designed to work well when the programmer is unsure what the program is supposed to do.
I don't see anything wrong with a language doing exactly what it was supposed to do.
Why? Do you have some facts that support your position?
First off most ATM software was not written in COBOL. However problems with COBOL
1) No anonymous data structures 2) No typing on points, essentially only a void pointer 2') Pointing to various types of dynamic or statically allocated structures is quite complex 3) Weak conditionals 4) No first class functions 5) No recursion 6) Limited variable constructors 7) No garbage collection 8) No obvious parse trees (yet another problem with build DSL in COBOL)
The thing is that the hashing computation is passed off to the client during backup:
imagine something like:
foreach b in (blocks in file)
compute h=hash(b) # client
if h not in (backedup blocks) # server
then backup b # server
else file = file (sub b) # client
or something. All that except for the query for which blocks are backed up (backup b)
Palm had a better OS Palm had a better UI for apps CE systems in general had better CPUs CE systems had more memory CE systems had better screens CE systems could import excel almost natively CE systems sinced better with outlook
I'd say that CE systems were "better" for what the majority wanted. And this is coming from a guy who bought a Palm during that time
Yes, I used something similar to ZFS for mass document storage a few years back. You do a complex checksum on the block level. Any two blocks with the same sum are the same. Each unique block is only stored once, though multiple files might link to it. You're right the file system doesn't know why you are using the same blocks over and over, but it doesn't care.
if i've got a bunch of files that take up 700mb on a ZFS device and try to back up to a (Joliet) CD will i get a message telling me that the CD doesnt have room?
Well if that's the case it makes the whole discussion about why buy a MMORPG more interesting. We have lots of good free engines and really what's needed are art guys which the open source community has had trouble recruiting.
With ZFS we might be able to get some very powerful backup features into OSX. Most binaries files don't change most of their content, ZFS makes it possible to due meaningful differential backups on large binary files. So for example 200 versions of a word doc with sounds and pictures that got revised over 6 months get stored in maybe 3x the space of the last revision. Emails with the same attachments get stored in just a few k rather than taking a meg each.... If Apple has this all working together by 10.5 then TimeMachine will work far far better then people currently expect it to. A 50g drive will be backing up a terabyte of worth of files.
Do you really think that the price of an airport hamburger is decided by the price of making it? Or if you buy a sandwich on the flight (many sell them now) that the price up there is determined by how much it costs to manufacture?
Yeah the cost of the airport hamburger includes the cost to the restaurant of renting space in the airport. The cost of the in-flight sandwich includes what it costs to running catering in the air (with security and enormous time constraints, weight being a big issue...).
I'd agree with that adjustment. My main point is that if we let
Yn = number of people interested in the product at $n per copy then Y0 is much much larger then Y(price). And so a model which distributes the fixed cost over Y(price) doesn't really reflect the reality of who would want the software for free / much cheaper.
In your situation the support costs are low (a user is only 1/1000 likely to need support at all). So really you have a situation where the first copy costs a fortune and latter copies cost nothing. The natural way to price that sort of thing would be for an entity interested in the market as a whole to pay you a large fixed cost and then the software to be free. Something like an Microsoft (interested in selling more OSes), Intel or the government. The software economy isn't set up like that and as an individual you can't create those sorts of social structures. That requires a political understand (i.e. changes in policy). As an individual you may very well have to utilize an artificial economy, that doesn't change what is the natural state for such an economy.
I suspect if your fixed costs are 100,000 you would need far more than 10,000 users paying $20 to pay for the app. Making the sale is going to be expensive. My guess is the numbers are closer to:
100k users at $10 each (with $7/copy going to sales expenses) or more likely: 10k users at $50 each ($10 / copy going to sales expenses)
And then you have all sorts of pricing problems. Because there may be millions of people who would have used their software if it were bundled and for them it might be worth $5 / copy. You just have no way to sell to them.
According to what theory does the price of production need to be reflected in the price of purchase?
Just about every economic model. If the cost of production is much higher than price then production stops, much lower and undue competition is created....
And that BTW is how Richard Stallman came up with the whole idea that software represented an artificial economy. When in actual practice a good costs: -- a lot to make the first copy of -- very little / nothing to make additional copies of -- a lot per copy for support
The obvious places to charge for the good are on support and initial development. A per copy charge is completely irrational given that price structure for production.
The Apple ][e's when fully loaded were 16k rom, 48k ram, 64k expanded memory (standard in ][c.) Anyway a high end 286 (the kind that was used for OS2 1.3 for example) might have had 4 megs. But going to 4 megs as part of a 286->386 upgrade would have been reasonable. Anyway if that's the case the time for the memory was a bios setting. You didn't have to live with it:-(. There were 3 settings: full, quick and none. You were probably doing a full check and yes that would have taken over a minute.
Where did you get the 8086 with 4 megs of ram? Especially since it could only address 1 meg (excluding expanded memory which were nowhere near that big). 4 megs was pretty unusual even for 80286s (which were generally 1 meg) the low end 386s didn't have 4 megs until around the time of the 80486-33; because it was the arrival of the the 4 meg sims that dropped the price on 1 meg sims to a level where joe average buyer wanted enough ram to run more than Dos apps.
Where do you think that VBA code came from in the first place? VBA did its job back when that code was written, an office type power user was able to create a fully automated application for a business process usually in under 2 weeks. Those were never meant to be permanent apps.
Framemaker on Macs and Solaris machines were what Technical Writers used - period.
What year are you talking? TeX/LaTex came out around the same time as postscript and the whole move from VMS/Unix mini computer typesetting to PC typesetting (around the mid 1980s). The old stuff was TeX like (troff, XIX...). Framemaker (along with stuff like Ventura) was huge for long docs up until the early 1990s but the technical community has always been divided.
Anyway as far as LaTex goes there are lots of fairly easy Latex based solutions (do a search under TeX guis). As far as Word you have to get people to use chapters. t
In fact I think the spiled brattiness of Mac users are an actual discouragement for open source programmers.
What are you on drugs? 1/2 the open source programmers are Mac users. Most open source programmers are thrilled to have a full featured desktop with a high quality built in Unix. They love the Apple development environment and its excellent support for open source languages. Sun (which funds Star/Open Office) isn't that interested in Mac
Think about this for a second. Do you think the people who are interested in "the standard" rather than what they think is best would be using OSX at all? X is designed to work well for people who like Unix apps (Darwin users). Its also designed to offer some level of support for an integrated environment. But that's far short of a mac app.
Slashdot Mirror
I think when they say VB is OO they mean it the way people talk about Scheme being OO (that is it supports object design methods, that is data structures can have functions associated with them. I don't think they mean its OO in the sense that it uses the Java/Smalltalk paradigm of syntax. Further the LINQ direction seems to be to move more towards the Lambda family of languages; which given C#'s move in the opposite direction makes sense. Why should Microsoft only support one programming paradigm?
Who cares whether its a technology you would have chosen? What you have with the VB program is:
1) A fully working prototype
2) Business rules and logic spelled out and worked through
3) A GUI interface agreed to with the customer
4) Clear limitations of the above model understood.
5) An understand of how this app is going to mature overtime
Why doesn't that justify a rewrite at this point?
How much percent of actual professional VB developers is out there?
In 2007 a very low number. In 1995 a much higher number. In 1995 there was for all practical purposes no Java. GUIs were very platform specific and very complex requiring a great deal of understanding of underlying OS concepts. C/C++ required programmers to work as such a low level that business rules were obscured, hat is programmers were something like 15x as productive in VB as they were in the C++ GUI languages.
Corporate American had definitely moved towards the Windows desktop without thinking through a strategy for writing a huge quantity of client programs quickly enough that they wouldn't need a major round of mainframe / mini computer upgrades. In that world VB, Lotus/Excel, Foxpro/Access/Paradox programming comprised something like 70% of all development.
te a horrible unmaintenable mess which not even the original author will touch, and which has is almost certainly going to be rewritten by a developer at some point in future
That would actually be a perfect end for a VB program. After the iterative process of requirements gathering is done via a VB rapid development methodology it wouldn't be at all unreasonable to then come in and rewrite it. The rewriters now have:
1) A fully working prototype
2) Business rules and logic spelled out and worked through
3) A GUI interface agreed to with the customer
4) Clear limitations of the above model understood.
So what exactly went wrong?
Because that wasn't the intention. VB (VB 3-6 not .NET) apps were meant to be used for short term and limited goals. If a program is going to be used for an extended period of time and modified it shouldn't have been written in VB. Clarity of concept requires organization of concept. VB was designed to work well for an iterative process of discovering the organization, that is VB is designed to work well when the programmer is unsure what the program is supposed to do.
I don't see anything wrong with a language doing exactly what it was supposed to do.
Why? Do you have some facts that support your position?
First off most ATM software was not written in COBOL. However problems with COBOL
1) No anonymous data structures
2) No typing on points, essentially only a void pointer
2') Pointing to various types of dynamic or statically allocated structures is quite complex
3) Weak conditionals
4) No first class functions
5) No recursion
6) Limited variable constructors
7) No garbage collection
8) No obvious parse trees (yet another problem with build DSL in COBOL)
I could go on but that list works for now
The thing is that the hashing computation is passed off to the client during backup:
imagine something like:
foreach b in (blocks in file)
compute h=hash(b) # client
if h not in (backedup blocks) # server
then backup b # server
else file = file (sub b) # client
or something. All that except for the query for which blocks are backed up (backup b)
At the time when the CE wars were being faught:
Palm had a better OS
Palm had a better UI for apps
CE systems in general had better CPUs
CE systems had more memory
CE systems had better screens
CE systems could import excel almost natively
CE systems sinced better with outlook
I'd say that CE systems were "better" for what the majority wanted. And this is coming from a guy who bought a Palm during that time
I don't remember exactly. I think 4k, and checksum was something like 512 bits = 64 bytes (so it wasn't MD5). Why?
does some funky heuristic happen?
Yes, I used something similar to ZFS for mass document storage a few years back. You do a complex checksum on the block level. Any two blocks with the same sum are the same. Each unique block is only stored once, though multiple files might link to it. You're right the file system doesn't know why you are using the same blocks over and over, but it doesn't care.
if i've got a bunch of files that take up 700mb on a ZFS device and try to back up to a (Joliet) CD will i get a message telling me that the CD doesnt have room?
Assuming you have repetitive block, yes.
Well if that's the case it makes the whole discussion about why buy a MMORPG more interesting. We have lots of good free engines and really what's needed are art guys which the open source community has had trouble recruiting.
Daimon
With ZFS we might be able to get some very powerful backup features into OSX. Most binaries files don't change most of their content, ZFS makes it possible to due meaningful differential backups on large binary files. So for example 200 versions of a word doc with sounds and pictures that got revised over 6 months get stored in maybe 3x the space of the last revision. Emails with the same attachments get stored in just a few k rather than taking a meg each.... If Apple has this all working together by 10.5 then TimeMachine will work far far better then people currently expect it to. A 50g drive will be backing up a terabyte of worth of files.
Do you really think that the price of an airport hamburger is decided by the price of making it? Or if you buy a sandwich on the flight (many sell them now) that the price up there is determined by how much it costs to manufacture?
Yeah the cost of the airport hamburger includes the cost to the restaurant of renting space in the airport. The cost of the in-flight sandwich includes what it costs to running catering in the air (with security and enormous time constraints, weight being a big issue...).
I'd agree with that adjustment. My main point is that if we let
Yn = number of people interested in the product at $n per copy
then Y0 is much much larger then Y(price). And so a model which distributes the fixed cost over Y(price) doesn't really reflect the reality of who would want the software for free / much cheaper.
Where are we supposed to make money?
In your situation the support costs are low (a user is only 1/1000 likely to need support at all). So really you have a situation where the first copy costs a fortune and latter copies cost nothing. The natural way to price that sort of thing would be for an entity interested in the market as a whole to pay you a large fixed cost and then the software to be free. Something like an Microsoft (interested in selling more OSes), Intel or the government. The software economy isn't set up like that and as an individual you can't create those sorts of social structures. That requires a political understand (i.e. changes in policy). As an individual you may very well have to utilize an artificial economy, that doesn't change what is the natural state for such an economy.
I suspect if your fixed costs are 100,000 you would need far more than 10,000 users paying $20 to pay for the app. Making the sale is going to be expensive. My guess is the numbers are closer to:
100k users at $10 each (with $7/copy going to sales expenses) or more likely: 10k users at $50 each ($10 / copy going to sales expenses)
And then you have all sorts of pricing problems. Because there may be millions of people who would have used their software if it were bundled and for them it might be worth $5 / copy. You just have no way to sell to them.
Not really. Without owning a very expensive factory I can't duplicate a CPU (which is a great example). I'm not paying the first copy but rather
-- fixed costs (F) are very high. Allow for the production of Y
-- per copy costs are x
Price will tend towards: (F/Y+x) * m
where m is the profit margin. For software x is tiny and Y is infinite.
According to what theory does the price of production need to be reflected in the price of purchase?
Just about every economic model. If the cost of production is much higher than price then production stops, much lower and undue competition is created....
And that BTW is how Richard Stallman came up with the whole idea that software represented an artificial economy. When in actual practice a good costs:
-- a lot to make the first copy of
-- very little / nothing to make additional copies of
-- a lot per copy for support
The obvious places to charge for the good are on support and initial development. A per copy charge is completely irrational given that price structure for production.
The Apple ][e's when fully loaded were 16k rom, 48k ram, 64k expanded memory (standard in ][c.) Anyway a high end 286 (the kind that was used for OS2 1.3 for example) might have had 4 megs. But going to 4 megs as part of a 286->386 upgrade would have been reasonable. Anyway if that's the case the time for the memory was a bios setting. You didn't have to live with it :-(. There were 3 settings: full, quick and none. You were probably doing a full check and yes that would have taken over a minute.
Where did you get the 8086 with 4 megs of ram? Especially since it could only address 1 meg (excluding expanded memory which were nowhere near that big). 4 megs was pretty unusual even for 80286s (which were generally 1 meg) the low end 386s didn't have 4 megs until around the time of the 80486-33; because it was the arrival of the the 4 meg sims that dropped the price on 1 meg sims to a level where joe average buyer wanted enough ram to run more than Dos apps.
Where do you think that VBA code came from in the first place? VBA did its job back when that code was written, an office type power user was able to create a fully automated application for a business process usually in under 2 weeks. Those were never meant to be permanent apps.
Framemaker on Macs and Solaris machines were what Technical Writers used - period.
What year are you talking? TeX/LaTex came out around the same time as postscript and the whole move from VMS/Unix mini computer typesetting to PC typesetting (around the mid 1980s). The old stuff was TeX like (troff, XIX...). Framemaker (along with stuff like Ventura) was huge for long docs up until the early 1990s but the technical community has always been divided.
Anyway as far as LaTex goes there are lots of fairly easy Latex based solutions (do a search under TeX guis). As far as Word you have to get people to use chapters. t
In fact I think the spiled brattiness of Mac users are an actual discouragement for open source programmers.
What are you on drugs? 1/2 the open source programmers are Mac users. Most open source programmers are thrilled to have a full featured desktop with a high quality built in Unix. They love the Apple development environment and its excellent support for open source languages. Sun (which funds Star/Open Office) isn't that interested in Mac
X is a standard for windowing on *nix systems
Think about this for a second. Do you think the people who are interested in "the standard" rather than what they think is best would be using OSX at all? X is designed to work well for people who like Unix apps (Darwin users). Its also designed to offer some level of support for an integrated environment. But that's far short of a mac app.