According to the article, the hole in Subversion was "caused by an error in the way the code parses dates". Now, incorrect date parsing is just as easy to do in Java (or C# or Python or any non-memory-managing language) as it is in C/C++. I haven't read more detailed information about the Subversion hole, but if you think you're safe just because you're programming in a language/environment where buffer overflows can't happen, you've got another thing coming.
For a overview of the field that is full-text search (of which web search engines is an important, but not the only, part), you should read Tim Bray's essays on search. He's been working on full-text search for a long time, knows his stuff and explains it in a very readable manner.
Or you could use a decent RSS reader (one that implements If-modified-since) together with their RSS feed at http://www.groklaw.net/backend/GrokLaw.rdf
Actually, Microsoft is really serious about backwards compatibility, both binary and source. And they do tell 3rd party developers when their public API's change. Here's a list of changes from.Net 1.0 to 1.1, for example. http://www.gotdotnet.com/team/changeinfo/default.a spx
However, it's still possible to write applications that rely on undocumented behaviour of the public API's, and occassionally such apps will break when run on new platforms. MS does jump through some pretty serious hoops to keep apps running even when they're doing "wrong", though. Maybe the impression that new MS platforms break old 3rd party stuff comes from the fact that there are a hell of a lot of 3rd party stuff out there?
Raymond Chen's blog provides lots of details of how much thought goes into backwards compatibility. An interesting read: http://weblogs.asp.net/oldnewthing/
While a small team of good people almost always produce good output, I think XP will give much better results than a heavyweight methodology like RUP.
Conversely, a team of average or mediocre people using RUP will probably produce acceptable result, but if the same team uses XP (or another light-weight metodology) the project may very well crash and burn.
Eric Sink is well known for liking the.NET platform. Since he has actual experience developing fairly complex products with both.NET and Java, I don't think anyone should be quick to dismiss him as a Microsoft fanboy.
Re:The one i hate most
on
Software Fashion
·
· Score: 1, Flamebait
It's amazing how people keep missing my point. Of course I can look up the definitions. Modern IDE's/texteditors will even give me the type just by hovering my mouse over the name or in some similar fashion.
My point (reiterated in the two other comments in this thread) is that it's just too slow, especially when skimming through code you're not intimately familiar with. It's just faster to get a clear view of the code if the types can be deduced just by looking at the code directly in front of me, without hovering, switching to header files definitions, or grepping the source tree.
I've even suggested a way that would bring the benefit of this easy skimmability (is that a word?) without requiring hungarian notation (using colors/icons/static tooltips), but so far noone seems to know of such an editor/IDE.
Re:The one i hate most
on
Software Fashion
·
· Score: 1, Flamebait
And what is "num" in "numpoints" but a wordier, less typesafe form of hungarian notation? How can you be sure that "filename" is a char*, as opposed to a C++ string object, a BSTR, or a wchar_t*? Maybe you don't code in those environments, but I do. I need to be able deduce a variable's type, period. For now, the name is the quickest, most convenient way of doing that, as far as I know.
In response to your question, HN is at it's most useful when reading the code, before you decide what to actually do with it (as I explained in the post you commented). This needs to be done when first getting aquainted with the code, and often when editing a part that isn't in your brain cache as well.
Re:The one i hate most
on
Software Fashion
·
· Score: 2, Insightful
What was so bad with just looking up the damn variable's definition! (Rather easy with most modern IDEs)
I have not seen an IDE which makes it as painless to deduce a variables type quickly as hungarian notation (once you learn what "sz" "lp", "pI" and so on stands for). When scanning through new code this helps a lot, even if it's not done 100% consistently. Not even tooltips (as VS.NET and several other IDEs has) are quick enough.
I do think such an IDE (or editor, really) could be constructed, perhaps using different coloring, permanent tooltips or icons to indicate types, scope, access levels etc, and when I start using one I shall drop the usage of hungarian notation.
Does anyone know of such an IDE or editor? I would really like to know.
Some good points, but...
on
Software Fashion
·
· Score: 2, Interesting
These guys have some points, but I think a much better article could be written about this topic. In particular, I object to some statements made:
Robin [...]was once asked during a job interview: "What's your favourite design pattern?" What's the correct response to that?
I don't think that's such a stupid question, as long as it's interpreted correctly. A good design pattern, like a good algorithm, is likeable in its own right, because of it's elegance and the way it breaks down a complicated problem. Maybe the interviewer wanted to know if Robin was really passionate about programming.
VB.Net is really just syntactic sugar on top of C#. C# offers more and better libraries.
What libraries do C# offer that are not accessible from VB.NET? As far as I know, all C# libraries (at least those in the standard framework) are CLS compliant, and thus accessible from any CLR language.
Because programmers didn't test that much, XP stipulates that tests must be written before the code. In other words, just because something has a weakness you shouldn't do the opposite in an extreme form.
That's just crazy talk. Automated regression tests isn't intended to relieve those lazy programmers, in XP they're the de-facto definition of what the system is designed to. Not to mention that test-first design often leads to better design, in particular wrt coupling between classes and components.
Re:Java's not exactly pining for the fields just n
on
Java vs .NET
·
· Score: 1
(Yes, VS has its problems, but it has a lot of unique tools, like compile-and-continue, which save hours!)
While I do like the VS.net environment, I must point out that compile-and-continue is NOT supported for C#, which is considered the most "native".NET language. It is not supported for VB.NET either, which makes it a step backwards for VB programmers.
I found writing COM servers in C++ using ATL to be fairly painless (including stuff like implementing the ActiveX interfaces and IEnumerable). Certainly not as painless as doing it with VB, but not that bad.
Re: developer communities size, I find that there exists countless sites, blogs, newsgroups and mailinglist that cover all angles of.NET development.
Apart from MSDN, I often find answers to my questions at gotdotnet.com, codeproject.com, www.dotnet247.com and the microsoft newsgroups. Brad Adams' and Chris Brumme's are very helpful for the low-level aspects of the.NET framework.
AFAIK, every developer worth his salt dates everything
I let my revision control system (eg CVS) do that for me. That way, it doesn't clutter the code, unless I want it to (eg "cvs annotate"). Printouts of the kind described in the article would not have this information, though (and in Linux case it would not be available, as no revision control system was used until recently)
Depending on what kind of digital satellite you're getting, there might be a way of streaming the data directly to disc with MythTV. The new 0.9 version has preliminary DVB support, if you have such a card in your PC (and a similar project, DVR, is *built* for these cards). However, I think DVB is mostly used in Europe, and so it might not be an option for everyone.
The Nazi paralells were in the way the visitors slowly took over the planet. Remember; they did not take it over with brute force at first, but through manipulation, much like the Nazi party rose to power.
They fabricated a conspiracy in the scientific community, making those in power treat scientists (and later on, anyone related to or in contact with scientist) as criminals, paralelling the way the Nazi's made the german society mistrust the Jews.
They also made sure to control the mass media early on, telling their version of events (paralelling Goebbels quote "If you tell a big enough Lie, and keep on repeating it, in the end people will come to believe it")
Later on, scientists and other visitor enemies were placed in labour camps, which was in reality genocide camps, another paralell.
The visitor friends program was an obvious Hitlerjugend paralell.
In other aspect (such as the human informants), it more resembled East europe communism.
How could you forget MythTV [mythtv.org], particularly when Freevo is just a ripoff of MythTV source?
That's bullshit, Freevo and MythTV have completely separate codebases (Freevo is built using python + some C parts for display), MythTV is built on C++ and QT.
One thing that sucks with webdav integration in Win2K is that it looks like a normal network file system, yet it's not (not accessible from the command line, you can't just click on a webdav mounted mp3 and have it open in winamp, etc). Maybe this apple tool it better that way?
No, the primary issue is as ever, interoperability, as in:
"can it read word and power point files?"
"can it sync with my Outlook/Exchange system?"
"Can it read IE-specific websites (incl stuff like client-side scripting and NTLM auth)?"
"can I use it to connect to my LAN through my windows-based VPN?"
I'm not saying that these IBM/QTopia devices won't be able to do all of the above, but I'm pretty sure that the Pocket PC devices will make it a lot easier.
Is it possible to build something like this so my files can only be accessed, or at least can only be decrypted, by me, and hence are not usable to the person providing the disk space?
If you had read the DIBS introduction on the linked page, you would have seen the following: Note that DIBS is a backup system not a file sharing system like Napster, Gnutella, Kazaa, etc. In fact, DIBS encrypts all data transmissions so that the peers you trade files with can not access your data.
Slashdot Mirror
It's interesting to note that Anders Hejlsberg, who was behind much of Delphi and J++, is one of the inventors in this patent.
According to the article, the hole in Subversion was "caused by an error in the way the code parses dates". Now, incorrect date parsing is just as easy to do in Java (or C# or Python or any non-memory-managing language) as it is in C/C++. I haven't read more detailed information about the Subversion hole, but if you think you're safe just because you're programming in a language/environment where buffer overflows can't happen, you've got another thing coming.
For a overview of the field that is full-text search (of which web search engines is an important, but not the only, part), you should read Tim Bray's essays on search. He's been working on full-text search for a long time, knows his stuff and explains it in a very readable manner.
Or you could use a decent RSS reader (one that implements If-modified-since) together with their RSS feed at http://www.groklaw.net/backend/GrokLaw.rdf
Actually, Microsoft is really serious about backwards compatibility, both binary and source. And they do tell 3rd party developers when their public API's change. Here's a list of changes from .Net 1.0 to 1.1, for example. http://www.gotdotnet.com/team/changeinfo/default.a spx
However, it's still possible to write applications that rely on undocumented behaviour of the public API's, and occassionally such apps will break when run on new platforms. MS does jump through some pretty serious hoops to keep apps running even when they're doing "wrong", though. Maybe the impression that new MS platforms break old 3rd party stuff comes from the fact that there are a hell of a lot of 3rd party stuff out there?
Raymond Chen's blog provides lots of details of how much thought goes into backwards compatibility. An interesting read: http://weblogs.asp.net/oldnewthing/
While a small team of good people almost always produce good output, I think XP will give much better results than a heavyweight methodology like RUP.
Conversely, a team of average or mediocre people using RUP will probably produce acceptable result, but if the same team uses XP (or another light-weight metodology) the project may very well crash and burn.
The * character works as a wildcard when searching for strings. Try this one for some variations of a famous quote.
Eric Sink is well known for liking the .NET platform. Since he has actual experience developing fairly complex products with both .NET and Java, I don't think anyone should be quick to dismiss him as a Microsoft fanboy.
It's amazing how people keep missing my point. Of course I can look up the definitions. Modern IDE's/texteditors will even give me the type just by hovering my mouse over the name or in some similar fashion.
My point (reiterated in the two other comments in this thread) is that it's just too slow, especially when skimming through code you're not intimately familiar with. It's just faster to get a clear view of the code if the types can be deduced just by looking at the code directly in front of me, without hovering, switching to header files definitions, or grepping the source tree.
I've even suggested a way that would bring the benefit of this easy skimmability (is that a word?) without requiring hungarian notation (using colors/icons/static tooltips), but so far noone seems to know of such an editor/IDE.
And what is "num" in "numpoints" but a wordier, less typesafe form of hungarian notation? How can you be sure that "filename" is a char*, as opposed to a C++ string object, a BSTR, or a wchar_t*? Maybe you don't code in those environments, but I do. I need to be able deduce a variable's type, period. For now, the name is the quickest, most convenient way of doing that, as far as I know.
In response to your question, HN is at it's most useful when reading the code, before you decide what to actually do with it (as I explained in the post you commented). This needs to be done when first getting aquainted with the code, and often when editing a part that isn't in your brain cache as well.
What was so bad with just looking up the damn variable's definition! (Rather easy with most modern IDEs)
I have not seen an IDE which makes it as painless to deduce a variables type quickly as hungarian notation (once you learn what "sz" "lp", "pI" and so on stands for). When scanning through new code this helps a lot, even if it's not done 100% consistently. Not even tooltips (as VS.NET and several other IDEs has) are quick enough.
I do think such an IDE (or editor, really) could be constructed, perhaps using different coloring, permanent tooltips or icons to indicate types, scope, access levels etc, and when I start using one I shall drop the usage of hungarian notation.
Does anyone know of such an IDE or editor? I would really like to know.
These guys have some points, but I think a much better article could be written about this topic. In particular, I object to some statements made:
Robin [...]was once asked during a job interview: "What's your favourite design pattern?" What's the correct response to that?
I don't think that's such a stupid question, as long as it's interpreted correctly. A good design pattern, like a good algorithm, is likeable in its own right, because of it's elegance and the way it breaks down a complicated problem. Maybe the interviewer wanted to know if Robin was really passionate about programming.
VB.Net is really just syntactic sugar on top of C#. C# offers more and better libraries.
What libraries do C# offer that are not accessible from VB.NET? As far as I know, all C# libraries (at least those in the standard framework) are CLS compliant, and thus accessible from any CLR language.
Because programmers didn't test that much, XP stipulates that tests must be written before the code. In other words, just because something has a weakness you shouldn't do the opposite in an extreme form.
That's just crazy talk. Automated regression tests isn't intended to relieve those lazy programmers, in XP they're the de-facto definition of what the system is designed to. Not to mention that test-first design often leads to better design, in particular wrt coupling between classes and components.
(Yes, VS has its problems, but it has a lot of unique tools, like compile-and-continue, which save hours!)
.NET language. It is not supported for VB.NET either, which makes it a step backwards for VB programmers.
While I do like the VS.net environment, I must point out that compile-and-continue is NOT supported for C#, which is considered the most "native"
Wouldn't newer trojans compromise the metadata as well, as to hide their tracks?
I found writing COM servers in C++ using ATL to be fairly painless (including stuff like implementing the ActiveX interfaces and IEnumerable). Certainly not as painless as doing it with VB, but not that bad.
Dr. GUI on Components, COM, and ATL helped me a lot by covering what happens behind the scenes.
Re: developer communities size, I find that there exists countless sites, blogs, newsgroups and mailinglist that cover all angles of .NET development.
.NET framework.
Apart from MSDN, I often find answers to my questions at gotdotnet.com, codeproject.com, www.dotnet247.com and the microsoft newsgroups. Brad Adams' and Chris Brumme's are very helpful for the low-level aspects of the
AFAIK, every developer worth his salt dates everything
I let my revision control system (eg CVS) do that for me. That way, it doesn't clutter the code, unless I want it to (eg "cvs annotate"). Printouts of the kind described in the article would not have this information, though (and in Linux case it would not be available, as no revision control system was used until recently)
Depending on what kind of digital satellite you're getting, there might be a way of streaming the data directly to disc with MythTV. The new 0.9 version has preliminary DVB support, if you have such a card in your PC (and a similar project, DVR, is *built* for these cards). However, I think DVB is mostly used in Europe, and so it might not be an option for everyone.
The Nazi paralells were in the way the visitors slowly took over the planet. Remember; they did not take it over with brute force at first, but through manipulation, much like the Nazi party rose to power.
They fabricated a conspiracy in the scientific community, making those in power treat scientists (and later on, anyone related to or in contact with scientist) as criminals, paralelling the way the Nazi's made the german society mistrust the Jews.
They also made sure to control the mass media early on, telling their version of events (paralelling Goebbels quote "If you tell a big enough Lie, and keep on repeating it, in the end people will come to believe it")
Later on, scientists and other visitor enemies were placed in labour camps, which was in reality genocide camps, another paralell.
The visitor friends program was an obvious Hitlerjugend paralell.
In other aspect (such as the human informants), it more resembled East europe communism.
GSM upgrades easily to GPRS and GPRS-EDGE, which gives much of the 3G benefits, at a lower price.
For 3G alternatives, see also UTMS.
How could you forget MythTV [mythtv.org], particularly when Freevo is just a ripoff of MythTV source?
That's bullshit, Freevo and MythTV have completely separate codebases (Freevo is built using python + some C parts for display), MythTV is built on C++ and QT.
Since forever? It even validates!
One thing that sucks with webdav integration in Win2K is that it looks like a normal network file system, yet it's not (not accessible from the command line, you can't just click on a webdav mounted mp3 and have it open in winamp, etc). Maybe this apple tool it better that way?
I'm not saying that these IBM/QTopia devices won't be able to do all of the above, but I'm pretty sure that the Pocket PC devices will make it a lot easier.
If you had read the DIBS introduction on the linked page, you would have seen the following:
Note that DIBS is a backup system not a file sharing system like Napster, Gnutella, Kazaa, etc. In fact, DIBS encrypts all data transmissions so that the peers you trade files with can not access your data.