I'd round it out with a project based on public key (RSA, Elliptic Curves). Point them towards open sourced MP libraries and have them code their own implementation.
Or, have them write their own MP routines and then teach them about profiling and optimization.
Set up a web site for their efforts and then announce it on Freshmeat where others can get and examine their work. They'll learn a lot from just doing the project as well as what they can learn from others who extend or improve on their work. And, they'll become contributing members of the computing world as well. That alone should give them some satisfaction.
Interesting...the beta has been out for only a few months and already he has uptime stats for a year? Gotta get me one of those time machines!
In previous posts, somebody mentioned that there was little support for backups. Well, from what I learned last week, gbak allows for hot backups as well as transportability of databases between machines.
Replication technology is also available, so there should be little reason for any downtime next to a nuclear strike, IMHO.
Also, the performance stats run a few months ago were based upon the classic architecture rather than the super server architecture. I'd like to see those stats run again using the released version of IB 6. I think the results will be most interesting.
I've had IB 6 beta running on a lowly 486 for several months. Granted, it doesn't get exercised very frequently or heavily, but even this version is pretty stable and is always there when needed.
The documenation that is provided for earlier versions of Interbase (4 - 5.5) was actually pretty complete.
Since IB6 is a new animal from a new company, the effort seems to have been getting the program ready. Documentation seems a bit dated at the moment, but there seems to be an effort to remedy this deficiency.
There are references to sample code, utilites and support groups on the interbase website Interbase
Delphi 6 will include both the VCL and CLX libraries. You'll use the VCL for native Windows apps and CLX for cross platform development.
We've seen both the existing Kylix and Delphi 6. Both are not ready for shipping. When they are released, life will be good.
They seem to think Kylix will be ready for initial release this fall. D6 probably won't ship until sometime next year.
Yes, these delays were disappointing. We would have loved to come home with a copy of D6, Kylix, or even Open Sourced InterBase. Didn't happen. This is in sharp contract to the last two conferences where the stuff was near ready for shipping (or shipping). We got excited by it as we could play with it. Now, we just have to wait a little while longer. To paraphrase Dale Fuller (CEO, Inprise), "It won't ship until its ready.".
Borland is digging themselves out of a hole. No more major screwups. Rushing a product out the door might appease the stock holders, initially. But, when the problems surface, you can bet the stock price will drop. Borland doesn't want that to happen again and is taking the hit now...by not releasing prematurely.
For starters, Borland (yes, that's their name again...Borland...An Inprise company), is stressing Kylix as a tool for building business and end-user appliations for Linux. They are not even daring to say that it should be used for kernel or device driver development.
At the conference, one of Borland's team leaders (I can't remember which one at the moment), stress this point. They DON'T want to compete with GCC in this arena. Borland percieves a different market, one filled with developers who need to develop database-centric, cross-platform business or end-user applications.
The fear that so many are exhibiting becausing of the Kylix is absurb and simply uncalled for. Borland is aware of the concerns of the Open Source community. I suspect they have no intention on going the way of the dinosour by being bound to another vendor's OS. Hence, their tools will be cross platform.
True, tools like Delphi Enterprise or C++Builder Enterprise will run you about $2,500. But, I'd venture that the initial release of Kylix (whether it be the Delphi or C++Builder variant), will be substantially less priced than that as you can get Standard and Professional editions of their Windows counterparts at princes ranging from $99 to $400.
If using GCC is what makes someone a developer and not the promise to develop high quality applications in a fraction of the time, then I guess its time to seek a new profession as I must not be a real developer.
Yes, you can write outstanding code using GCC. You can write some pretty piss poor code as well. It's not the tool that makes the developer, but rather thee ability to leverage knowledge and know which tool is the right one for a particular job that defines the term developer.
How does one use a hammer to attach two pieces of wood? Why, with nails and hard blows. But, what if I wanted to use screws instead of nails? Do I use a hammer? No, I use a power screwdriver.
If I want to write handcrafted C or C++ code to extend the kernel, then GCC is one of the tools that will get me there. Similarly, if I need to develop a business application, I'll use a tool like Delphi, C++Builder, or Kylix (but not VB) simply because they are designed for the purpose of creating database-centric business and end-user applications.
I would not, under 99% of circumstance, consider using those tools for kernel work (nor Kylix). That is not their role. Even Borland makes it clear that kernel development is not the role of Kylix.
So, stop belittling those that develop using other tools. Maybe they KNOW something you don't about getting the job done in an expeditious and cost effective manner without sacrificing quality or performance.
Oh yeah..you might want to look upwards...see that bright object in the sky...It a cluster of tools like Kylix that will make you extinct.
If Borland's Kylix is anything like Delphi, it will have all the Experts (Wizards to MS fans) and code completion that you speak of.
BTW, Kylix will be Borland's Delphi/C++ Builder environment for Linux (now in beta).
The only drawbacks I can think of going to Kylix are its dependencies on QT. But, supposedly, they will allow those libs to be distributed..and are encouraging inclusion on Linux distributions.
The other drawback is that it is a commercial product. That won't stop me (I'm a corporate (and hobby) Delphi developer), but it may stop some people. Me, I just want my apps to port with the littlest amount of work and then to develop natively with the same interface I've come to enjoy.
Me thinks you are a bit out of date. Borland has made the best Pascal/RAD environments since the the early 90's. The limitations you experienced were probably due to the sixteen bit POS architecture the code had to run on.
But, from it's initial release, Delphi has allowed developers to produce Windows applications that blew VB apps away (which was its intended goal). Delphi was touted as a RAD tool. RAD, as we know, does not necesarily translate to better code...but rather quicker development cycles. Quicker development cycles results in happier clients.
Additionally, even the earliest Delphi (D1) produced compiled code that executed significantly faster than VB, offered true object orientation (rather than object based) and delivered on its promise. Why else would Microsoft go out of their way to hire away Borland Delphi team members and their chief architect if they didn't perceive it as threat?
The only major and recognized failing in Delphi is the Borland Database Engine (BDE). The BDE, was obsolete almost to the day it was introduced. In later versions, Borland abstracted away the BDE and made us all very happy. I now have native access to Oracle and InterBase as well as ADO, ODBC, and DAO access to a boat load of others (SYBASE).
Middleware tools are also there. Pick COM, CORBA or one of the other technologies (I personally use ASTA) to build database/distributed apps that run over the internet. Or choose between HTML and/or XML. It's all there.
Regarding no tools for Linux - Borland demonstrated their C++ compiler for Linux last year at their conference in Philadelphia. And, they also released JBuilder for Sun and Linux. Now, Kylix (Delphi (and C++???) and for Linux) and Delphi 6 loom on the horizon. If they are source code compatible (as we are expecting them to be), it will mean cross platform applications for Linux and that other OS.
I can go on for hours about the merits of Delphi (I used to code C/C++ but gave it up for Delphi). But, when push comes to shove, a Delphi developer will crank out a finished (and polished) windowed application significantly faster than most other tools.
Oh..and if you still doubt the quality of Borland tools, look at the reviews Delphi received in the past year. So, before you blast Borland tools, get your facts straight and don't base your opinions on something that existing eight years ago when all we had were 486's and an operating system called Windows 3.1.
This is a problem I'm running into as well as I have an idea for a product (to be commercial, actually) and the issue is who own my idea to the product and any code I generate.
The answer to who owns intellectual property is not always what you expect.
Let's assume the original work belongs to him. Now, the company want him to update it. Is he a salaried employee? What are the laws in his state regarding intellectual property developed while a salaried employee? Unforunately, each state is different in this regard.
Talk to good IP lawyer. The FSF may be able to help you as well.
Well, I shave at Mach 3. Talk about serious razor burn.
Imagine the windburn on Trixie's face as Speed goes down the highway at Mach 5!!
The fact is, the vehicle was named Mach 5 (like the razor named Mach 3). This does not imply the vehicle travelled at mach 5 nor that I can shave at Mach 3.
My QM is a bit rusty (over 12 years now since I last used it in college). But, one thing I do remember is the demonstration of how QM based equations, when taken to their limits (i.e. nullifying the minor terms), resolve to their Newtonian counterparts. It's only on the quantum level that these other term become significant.
The second thing I distinctly remember is how, when applying QM to relativity, it was shown that a moving mass has the ability to "bend/drag" space in such a manner as to exhibit a "force" equivalent to that of gravity.
This would imply that "gravitons" don't truly exist, but rather are created to explain this "gravity" effect of moving particles. And, a pointed out previously, unless you are at absolute zero, every particle has momenentum (or a non-zero velocity).
It also implies that General Relativity has not been compromised by QM. Kind of reassuring, don't you think?
Now, I don't keep up on my QM (and my knowledge is a bit dated), but I found these two lectures particularly illuminating.
As a writer of both commercial (aka proprietary) and open source software, I can tell you that if I were purchasing software for its security related features, I'd like to see how these features are implemented.
Having written communications programs that implement various encryption algorithms and protocols, I am amazed and awed that even RSA Data Security (now Security Dynamics) can get it wrong sometimes. That's why they publicly post the Public Key Standards and allow others to review and provide constructive critism to close the holes.
Same goes for software. Yes, open source does make finding weaknesses easier. But, it also makes it easier to close those holes as you have many eyes looking at it. Security through Obscurity does not work. The crypto community has realized this years ago.
Where does someone like the author of that article get off contradicting proven methodologies when its clear they have no understanding the process itself?
Any recruiter will tell you they have a abundance of high-paying positions to fill using Java or C++, but I have found a few who have even heard of Delphi, and they have been for low-paying positions.
This effect is primarily because most IT shops are run by either Microsoft lemmings who read PC Maganize for mangement advise or Unix shops. The Unix shops tend to use C++ and Java because no real RAD tools exist in those arenas. If you want to write for Unix, you best be using C++.
The fact that object oriented programming is so predominent in the Unix world should make Kylix a useful tool. The synergy of C++ and Kylix should create some exciting opportunities (Kylix is going to use (supposedly).so files that should be able to link with those produced by other tools).
So, let the C++ gurus do their thing and let the Kylix developers write the business apps and front ends. This can only be a winning combination.
Delphi jobs are particularly abundant in Texas as well. One of the Delphi tool vendors I work is located in Boise, Idaho. There are quite a few in the Boston area that use Delphi as well.
In the Philadelphia area, where I live, the jobs are primarly custom software houses or consultanting firms. But, a they also tend to be in the pharaceutical and accounting areas and games arena as well. The consultant agencies which I have spoken with tell me that they are constantly hiring and there is no lack of work now or in the forseeable future.
Judging by the people I speak with and the apparent total ignorance of Delphi by many developers tells me a few things...
1) Most IT or marketing shops are contracting for work being done in Delphi and probably don't know it. All they know is the apps come on or ahead of schedule and work.
2) I can still command my salary as their isn't a glut of Delphi (or Kylix) programmers yet.
3) Kylix developers will be in demand as more an more companies decide to support Linux. This will help raise my salary if I should decide to leave my current position.
4) More opportunities will be created as more startups seeks to fill the void of mission critical business applications.
Kewl.
Now, of Borland fails to deliver, then none of this will hold true and I will be forced to switch back to C++.
I think you are way out of line here. First, Borland/Inprise aren't insinuating that the other tools aren't professional. They are making the point that their tools are now ready for "professional" developers to use in a Linux environment.
As for you statement regarding hiring a programmer...I have been using Delphi since its introduction. Yes, jobs are far fewer thanks to the misguided managment who think Microsoft is the way to go. I got my job (and am being paid VERY well) BECAUSE I have such extensive Delphi skills. Many IT shops, once they realize the limitations of VB and the difficulty of C++ in getting robust apps out into the marketplace, switch to Delphi as their secret weapon.
Delphi is predominantly a West Coast and European tool. So, unless you're in those regions, you probably don't see many Delphi jobs adverstised.
BTW, you're so called "Gold Standard of Skill" is a bit elitist. Yes, its nice to know whats going on under the covers (that's what component developers do). But, having a developer who can take the tools I give them and mold them into a production quality app is more important to me (unless I want a component developer).
First, you need to upgrade to D5 (forget D4). The advantages are numerous. D2 was a kludge (like D4) and should never have been released. D3 and D5 are the versions that can be considered "stable" for real development work.
That aside, you do make some valid points.
#1 - Syntax: The verbositiy is inherent of Pascal. But, I don't see how this makes it difficult to read. Good clean Pascal coding makes it possible to write code that is very legible and easy to comprehend. It not the diffulty in reading the syntax that makes C++ programmers uncomfortable - its the lack of terseness and the need to to type more to say the same thing.
#2 - Liguistic flaws. Okay...I'll byte (pun intended) and give you this one. I agree, having to declare everything up front can make life a bit difficult. How many times I wish I had the ability to create a temporary variable in a block of code. To get around this, I use another function or procedure or redesign my code. The optimizer solves my problem while making the code, ultimately more readable (in my book).
#3 - Lack of guaranteed destruction. Delphi will clean up anything it creates on the stack except for variables it creates in the heap. Thus, objects created aren't destroyed because they reside in the heap. A simply try/finally block solves that problem quite handily. But, also destruction of the object that "owns" the object will clean up any contained objects as well. This means that when an application terminates, all objects are automatically freed. The caveat here is that the objects know how to free up any memory they allocate in their destructor or there will be memory leaks.
#4 - Genericity. This is where I do have to concede. Templates don't exist in Delphi at the present time. But, there are are generic container objects and untyped variables (ick!).
Supposedly, Kylix will introduct real macro support - a feature missing from earlier variants of Delphi. I don't know how extensive the macro processing will be...we'll have to wait and see.
One other thing that is missing from Delphi, in general, is multiple inheritence. I don't know if this is a blessing or a hinderance. But, much of this is possible using interfaces which ARE being ported to Kylix.
Yes, I am a Delphi developer. Been so since Delphi 1. Prior to that, I was a C/C++ and VB developer (more C/C++ with VB as "glue"). That was 1994 when Delphi was introduced. And,I don't program in C/C++ any longer (except on Linux...where I prefer Perl for most things). That is because I can develop 24x7 database and internet apps on a Windows platform using Delphi (well..assuming the OS can stay up 24x7).
I find the language and tool well suited for group development in addition to single developer RAD sessions. RAD, in my book, is fine for prototypes...but it generates some nasty looking code that is often hard to follow. Nothing is more important then getting down the concepts and then doing a real design.
The limitations you mention are pretty much specific to the difference between C++ and Delphi. Delphi developers can probably put together a list of things missing from C++ as well. Its all subjective and personal taste. A truly good programmer learns to adapt and overcome. They incorporate the best of what they've learned in other languages and use that knowledge whereever pssible.
I was a bit disappointed upon hearing that it was not open to the general public as well. But, that is what I get for not being a certified Borland tools vendor. Our company hopes to apply for that status in the near future (we've got clients (potential and actual) who have had enough of Microsoft. They want our product and they want it on Linux. Period.).
It's unfortunate that people tend to take these notices out of context and put a negative spin on them. I'm not sure they even belong here without some sort of serious explanation. But, there were people responding to this article who were asking questions like "What is Kylix anyway?". A simple link to community.borland.com would have answered that in a heartbeat.
For those that do participate in the tools program, they are elgible to receive the tools well in advance of the general public. The idea here, as you pointed out, is so that they can port their wares in time for the general release.
Truth be told, Borland is trying to release as much information as possible to help prepare Windows developers for the daunting task of taking on Linux. Linux ain't Win98 (thankfully!). And, it has its own quirks that Windows developers simply can't appreciate until they try it.
Delphi component and tools builders will have a tougher go at it mainly because it is not Windows and doesn't provide the Win32 API. Code that relied on low-level features of Windows simply won't work any longer without redesign.
Application developers won't have to worry about these issues as they use generally don't code to the API level. They are more likely to use the tools created by the component writers (who already dealt with those issuer).
Okay, so I'm digressing a bit. Take the notice for what it is...marketing hype targetted to Delphi developers to get them psyched for the imminient release of the one of the most significant advances for Windows developers.
I would have to agree with you on most points. However, it's a shame that you feel like you'll be flamed for saying you'd switch back to Pascal.
As a Delphi developer since Delphi 1 (who USED to code in C++), I wouldn't consider switching back. True, the official PASCAL standard is pretty poor. But, since Borland Pascal has become the defacto standard (at least in the Windoze world), they have extended the language considerably.
These extensions, coupled with the rich IDE and visual component library (VCL) have enabled developers like myself to develop hiqh quality mission critical 24x7 apps (really). Try that with the likes of VB. Face it, VBers, a serious internet/database app probably can't be done that can be run in 24x7 environment.
Why? Delphi provides true exception support, produces true compiled code that doesn't need a runtime environment. And, it can call every Windoze API function in a clear, concise manner. Even when Windows API calls go South, Delphi can detect and correct most of them. C++ programmers would find themselves at home here.
Database access is extremely rich as well. Out of the box Delphi Enterprise supports, Oracle, DB2, Paradox, DBase, InterBase, Sybase, ADO and ODBC (to name a few). MIDAS and Visibroker make it easy to write distributed apps that can use COM, DCOM, TCP/IP, or CORBA. Extensions to MIDAS make it easy to build XML-based database applications (Internet Express). And, there are thousands of 3rd party add-ons.
Where Delphi gets its bad rep is (in order of highest to lowest precedence):
a) Previously poor corporate management (it's getting better...really).
b) Corporate IT Management that obtains its IT knowledge from rags like PCMagazine (owned, I believe by MS).
c) Developers who jumped on the VB bandwagon because of job availability and salary (see (b)) who then become Microsoft drones.
d) Initial cost of VB was lower than Delphi (now rectified). Newbie developers purchased VB for home and then went looking for jobs (see (c)).
There are some features notably missing in the language (as they are in all languages). I miss operator overloading in particular. Multi-inheritence is something that I wish for on a very rare occurance. But, when push comes to shove and I need to write a commercial quality application that gives me high performance in the shortest amount of time, I'll go with Delphi over C++ (Borland C++Builder included) just about anytime.
The fact that it is now coming to Linux only means more opportunities for developers like myself. I'd love to see Linux on every desktop. My lowly 400mhz workstation boots into Gnome (that'll have to change) in two seconds. I haven't shut my Linux box down in over a month with no ill effects. As long as I can still build console apps and daemons using Kylix, I'll give up GNU too.
Where C++developers/GNU purists get the leg up is that C++ is pretty much an open standard. And, there are good, solid, free tools out there as well to make development possible. But, unless you don't care about performance, executable size, and compile speed, all you gain by staying with open source compilers is just that...free or opened source tools (for now, at least).
If the commercial tools compile the same code (without changes) as your free tools can and you get added benefits from doing so, then why not? Am I missing something to say if the BC++ compiler can compile the kernel and its smaller and faster, is that not a good thing?
The Delphi community has already adopted the Open Source philosophy. I suggest you go to Project JEDI and check out the JEDI Project. Jedi uses the MPL.
Although other OSI project for Delphi exist (most notably, Winshoes (an internet component suit), most of the projects current adopted by JEDI deal with conversion of Microsoft Windows API headers into forms usable by Delphi developers.
As a side note, Kylix (Delphi on Linux) has been demonstrated in Spain (early betas). Supposedly, the demo compiled 45K lnes of X-VCL in 2 seconds.
A port of BCC 5.5 to linux already exists. This was demonstrated at the 1999 Borland/Inprise Conference. Yes, the version released is for Windows. But, they demonstrated that they could run the same source through their Linux compiler, have it compile, and run as expected.
Borland is planning to release Kylix which includes (to my understanding) Delphi on Linux as well as C++ Builder for Linux.
If anyone has used their tools, you will find that they provide all the source code to the libraries (mainly the VCL). They don't include the source to the compilers or other tools (they need to make money somehow, right? Isn't that what for-profit corporations do?).
Nobody is forcing you to download the BCC. Nor, are they forcing you to visit their site. The fact that you went there says you are interested.
The only reasons I can think that they want to know who is downloading their compiler (i.e registration) and filling in the survey are:
1) Knowing if their marketing campaign is having an effect.
2) Being able to notify registered users when Kylix becomes available.
In a previous post, somebody griped about the BCC compiler stating that the Cygnus tools (even MS's own were preferable to BCC). The only downside I ever experienced using BCC was that many libraries in the Windows world simply would not link to it. But, I experienced the same problem with MS's and Symmantec's compilers as well. At least in the Linux world, there is a standard on how these libraries should be built. The Borland compilers should be compatible with other libraries on Linux. No?
So, why does everyone keep knocking Inprise/Borland for their approach of giving away the product?
Personally, it seems to me that too many people don't want to look at the good that can come from this. They expect everything to be free or a threat to their purist, unrealistic, view of the world. That's great for hobbiests. It doesn't make money. Or, perhaps they are staunch MS supporters who don't want to see Inprise/Borland succeed? Just read their posts and decide for yourself.
With PGP, and no doubt many other encryption schemes, this would not prove anything. The encryption program chooses a random session key to encrypt the data, and encrypts this session key with the user's key.
Your hypothesis is flawed (as was my original statement...I was assuming a sysmmetric cipher)and this would not thwart the detection mechanism. If you are forced to turn over your key AND the key you give is bogus, one of two things will occur:
a) They unencrypted document will contain garbage (I don't think PGP will even let you decrypt a document with a bogus or wrong key). End Result - they nail you for contempt.
b) Any digital signatures on the document will not be verified. End Result - they nail you for contempt.
In 99.9% of public key systems, the document is encrypted first with a symmetric algorithm (like CAST, IDEA, DES, etc) and the key encrypted with the receipients public key.
As the symmetric key is recovered during the decryption phase and the document is decrypted, supplying a bogus key won't help you any.
Actually, this possible using a technique like Winnowing and Chaffing.
W&C enables more than one message to be stored in an encrypted file. Additionally, garbage bytes are thrown in to confuse the issue even more. Only the intended receiver of a particular message will receive the plain text.
There was really good article on this in DDJ a couple of months ago.
Funny you mention the D.O.S. attacks and Internet filtering software.
It is a sad state of affairs to realize that we, the internet community, are being targeted by laws that we, ourselves, help foster.
Truth be told, cyber-business is here to stay. Laws will be enacted world-wide to curb the activities that a few regengades are engaged in. Billions of dollars were lost this week by companies and stock-holders in the US alone by the fear these individuals have created regarding the security of the internet.
Dot.com businesses are, in many cases, little guys who had an idea and used the internet to capitialize on it. Potentially, many will lose their business because of this so called "cyber-vandalism".
We (the internet community) are attacking our own resulting in restricted access, increased police activity and legislation to deal with these miscreants of cyber-society.
Don't be suprised when measures are put into effect at the backbone level to detect these types of attacks and terminate all access to the internet by individuals and/or ISP who engage in or allow such activity. What about entire countries being denied access? It will happen, eventually. Look at China.
Furthermore, it can be proven that you committed the crime. All they would need to do is encrypt your "plaintext" version with the key you supplied and compare it with the message they are holding. If the don't match and you don't turn over the real key, they can hold you in contempt at the very least.
In the US, there was a case a few years ago where a suspected child pornographer had encrypted his pictures. The prosecuters wanted him to hand over the decryption keys. He refused on the grounds it violated his rights against self incrimination. Wonder how that one ever turned out....
Slashdot Mirror
I'd round it out with a project based on public key (RSA, Elliptic Curves). Point them towards open sourced MP libraries and have them code their own implementation.
Or, have them write their own MP routines and then teach them about profiling and optimization.
Set up a web site for their efforts and then announce it on Freshmeat where others can get and examine their work. They'll learn a lot from just doing the project as well as what they can learn from others who extend or improve on their work. And, they'll become contributing members of the computing world as well. That alone should give them some satisfaction.
RD
Interesting...the beta has been out for only a few months and already he has uptime stats for a year? Gotta get me one of those time machines!
In previous posts, somebody mentioned that there was little support for backups. Well, from what I learned last week, gbak allows for hot backups as well as transportability of databases between machines.
Replication technology is also available, so there should be little reason for any downtime next to a nuclear strike, IMHO.
Also, the performance stats run a few months ago were based upon the classic architecture rather than the super server architecture. I'd like to see those stats run again using the released version of IB 6. I think the results will be most interesting.
I've had IB 6 beta running on a lowly 486 for several months. Granted, it doesn't get exercised very frequently or heavily, but even this version is pretty stable and is always there when needed.
RD
Since IB6 is a new animal from a new company, the effort seems to have been getting the program ready. Documentation seems a bit dated at the moment, but there seems to be an effort to remedy this deficiency.
There are references to sample code, utilites and support groups on the interbase website Interbase
You will. It's called Delphi 6.
Delphi 6 will include both the VCL and CLX libraries. You'll use the VCL for native Windows apps and CLX for cross platform development.
We've seen both the existing Kylix and Delphi 6. Both are not ready for shipping. When they are released, life will be good.
They seem to think Kylix will be ready for initial release this fall. D6 probably won't ship until sometime next year.
Yes, these delays were disappointing. We would have loved to come home with a copy of D6, Kylix, or even Open Sourced InterBase. Didn't happen. This is in sharp contract to the last two conferences where the stuff was near ready for shipping (or shipping). We got excited by it as we could play with it. Now, we just have to wait a little while longer. To paraphrase Dale Fuller (CEO, Inprise), "It won't ship until its ready.".
Borland is digging themselves out of a hole. No more major screwups. Rushing a product out the door might appease the stock holders, initially. But, when the problems surface, you can bet the stock price will drop. Borland doesn't want that to happen again and is taking the hit now...by not releasing prematurely.
For starters, Borland (yes, that's their name again...Borland...An Inprise company), is stressing Kylix as a tool for building business and end-user appliations for Linux. They are not even daring to say that it should be used for kernel or device driver development.
At the conference, one of Borland's team leaders (I can't remember which one at the moment), stress this point. They DON'T want to compete with GCC in this arena. Borland percieves a different market, one filled with developers who need to develop database-centric, cross-platform business or end-user applications.
The fear that so many are exhibiting becausing of the Kylix is absurb and simply uncalled for. Borland is aware of the concerns of the Open Source community. I suspect they have no intention on going the way of the dinosour by being bound to another vendor's OS. Hence, their tools will be cross platform.
True, tools like Delphi Enterprise or C++Builder Enterprise will run you about $2,500. But, I'd venture that the initial release of Kylix (whether it be the Delphi or C++Builder variant), will be substantially less priced than that as you can get Standard and Professional editions of their Windows counterparts at princes ranging from $99 to $400.
RD
If using GCC is what makes someone a developer and not the promise to develop high quality applications in a fraction of the time, then I guess its time to seek a new profession as I must not be a real developer.
Yes, you can write outstanding code using GCC. You can write some pretty piss poor code as well. It's not the tool that makes the developer, but rather thee ability to leverage knowledge and know which tool is the right one for a particular job that defines the term developer.
How does one use a hammer to attach two pieces of wood? Why, with nails and hard blows. But, what if I wanted to use screws instead of nails? Do I use a hammer? No, I use a power screwdriver.
If I want to write handcrafted C or C++ code to extend the kernel, then GCC is one of the tools that will get me there. Similarly, if I need to develop a business application, I'll use a tool like Delphi, C++Builder, or Kylix (but not VB) simply because they are designed for the purpose of creating database-centric business and end-user applications.
I would not, under 99% of circumstance, consider using those tools for kernel work (nor Kylix). That is not their role. Even Borland makes it clear that kernel development is not the role of Kylix.
So, stop belittling those that develop using other tools. Maybe they KNOW something you don't about getting the job done in an expeditious and cost effective manner without sacrificing quality or performance.
Oh yeah..you might want to look upwards...see that bright object in the sky...It a cluster of tools like Kylix that will make you extinct.
RD
If Borland's Kylix is anything like Delphi, it will have all the Experts (Wizards to MS fans) and code completion that you speak of.
BTW, Kylix will be Borland's Delphi/C++ Builder environment for Linux (now in beta).
The only drawbacks I can think of going to Kylix are its dependencies on QT. But, supposedly, they will allow those libs to be distributed..and are encouraging inclusion on Linux distributions.
The other drawback is that it is a commercial product. That won't stop me (I'm a corporate (and hobby) Delphi developer), but it may stop some people. Me, I just want my apps to port with the littlest amount of work and then to develop natively with the same interface I've come to enjoy.
RD
Me thinks you are a bit out of date. Borland has made the best Pascal/RAD environments since the the early 90's. The limitations you experienced were probably due to the sixteen bit POS architecture the code had to run on.
But, from it's initial release, Delphi has allowed developers to produce Windows applications that blew VB apps away (which was its intended goal). Delphi was touted as a RAD tool. RAD, as we know, does not necesarily translate to better code...but rather quicker development cycles. Quicker development cycles results in happier clients.
Additionally, even the earliest Delphi (D1) produced compiled code that executed significantly faster than VB, offered true object orientation (rather than object based) and delivered on its promise. Why else would Microsoft go out of their way to hire away Borland Delphi team members and their chief architect if they didn't perceive it as threat?
The only major and recognized failing in Delphi is the Borland Database Engine (BDE). The BDE, was obsolete almost to the day it was introduced. In later versions, Borland abstracted away the BDE and made us all very happy. I now have native access to Oracle and InterBase as well as ADO, ODBC, and DAO access to a boat load of others (SYBASE).
Middleware tools are also there. Pick COM, CORBA or one of the other technologies (I personally use ASTA) to build database/distributed apps that run over the internet. Or choose between HTML and/or XML. It's all there.
Regarding no tools for Linux - Borland demonstrated their C++ compiler for Linux last year at their conference in Philadelphia. And, they also released JBuilder for Sun and Linux. Now, Kylix (Delphi (and C++???) and for Linux) and Delphi 6 loom on the horizon. If they are source code compatible (as we are expecting them to be), it will mean cross platform applications for Linux and that other OS.
I can go on for hours about the merits of Delphi (I used to code C/C++ but gave it up for Delphi). But, when push comes to shove, a Delphi developer will crank out a finished (and polished) windowed application significantly faster than most other tools.
Oh..and if you still doubt the quality of Borland tools, look at the reviews Delphi received in the past year. So, before you blast Borland tools, get your facts straight and don't base your opinions on something that existing eight years ago when all we had were 486's and an operating system called Windows 3.1.
RD
This is a problem I'm running into as well as I have an idea for a product (to be commercial, actually) and the issue is who own my idea to the product and any code I generate.
The answer to who owns intellectual property is not always what you expect.
Let's assume the original work belongs to him. Now, the company want him to update it. Is he a salaried employee? What are the laws in his state regarding intellectual property developed while a salaried employee? Unforunately, each state is different in this regard.
Talk to good IP lawyer. The FSF may be able to help you as well.
Well, I shave at Mach 3. Talk about serious razor burn.
Imagine the windburn on Trixie's face as Speed goes down the highway at Mach 5!!
The fact is, the vehicle was named Mach 5 (like the razor named Mach 3). This does not imply the vehicle travelled at mach 5 nor that I can shave at Mach 3.
My QM is a bit rusty (over 12 years now since I last used it in college). But, one thing I do remember is the demonstration of how QM based equations, when taken to their limits (i.e. nullifying the minor terms), resolve to their Newtonian counterparts. It's only on the quantum level that these other term become significant.
The second thing I distinctly remember is how, when applying QM to relativity, it was shown that a moving mass has the ability to "bend/drag" space in such a manner as to exhibit a "force" equivalent to that of gravity.
This would imply that "gravitons" don't truly exist, but rather are created to explain this "gravity" effect of moving particles. And, a pointed out previously, unless you are at absolute zero, every particle has momenentum (or a non-zero velocity).
It also implies that General Relativity has not been compromised by QM. Kind of reassuring, don't you think?
Now, I don't keep up on my QM (and my knowledge is a bit dated), but I found these two lectures particularly illuminating.
RD
As a writer of both commercial (aka proprietary) and open source software, I can tell you that if I were purchasing software for its security related features, I'd like to see how these features are implemented.
Having written communications programs that implement various encryption algorithms and protocols, I am amazed and awed that even RSA Data Security (now Security Dynamics) can get it wrong sometimes. That's why they publicly post the Public Key Standards and allow others to review and provide constructive critism to close the holes.
Same goes for software. Yes, open source does make finding weaknesses easier. But, it also makes it easier to close those holes as you have many eyes looking at it. Security through Obscurity does not work. The crypto community has realized this years ago.
Where does someone like the author of that article get off contradicting proven methodologies when its clear they have no understanding the process itself?
RD
This effect is primarily because most IT shops are run by either Microsoft lemmings who read PC Maganize for mangement advise or Unix shops. The Unix shops tend to use C++ and Java because no real RAD tools exist in those arenas. If you want to write for Unix, you best be using C++.
The fact that object oriented programming is so predominent in the Unix world should make Kylix a useful tool. The synergy of C++ and Kylix should create some exciting opportunities (Kylix is going to use (supposedly) .so files that should be able to link with those produced by other tools).
So, let the C++ gurus do their thing and let the Kylix developers write the business apps and front ends. This can only be a winning combination.
RD
Delphi jobs are particularly abundant in Texas as well. One of the Delphi tool vendors I work is located in Boise, Idaho. There are quite a few in the Boston area that use Delphi as well.
In the Philadelphia area, where I live, the jobs are primarly custom software houses or consultanting firms. But, a they also tend to be in the pharaceutical and accounting areas and games arena as well. The consultant agencies which I have spoken with tell me that they are constantly hiring and there is no lack of work now or in the forseeable future.
Judging by the people I speak with and the apparent total ignorance of Delphi by many developers tells me a few things...
1) Most IT or marketing shops are contracting for work being done in Delphi and probably don't know it. All they know is the apps come on or ahead of schedule and work.
2) I can still command my salary as their isn't a glut of Delphi (or Kylix) programmers yet.
3) Kylix developers will be in demand as more an more companies decide to support Linux. This will help raise my salary if I should decide to leave my current position.
4) More opportunities will be created as more startups seeks to fill the void of mission critical business applications.
Kewl.
Now, of Borland fails to deliver, then none of this will hold true and I will be forced to switch back to C++.
RD
I think you are way out of line here. First, Borland/Inprise aren't insinuating that the other tools aren't professional. They are making the point that their tools are now ready for "professional" developers to use in a Linux environment.
As for you statement regarding hiring a programmer...I have been using Delphi since its introduction. Yes, jobs are far fewer thanks to the misguided managment who think Microsoft is the way to go. I got my job (and am being paid VERY well) BECAUSE I have such extensive Delphi skills. Many IT shops, once they realize the limitations of VB and the difficulty of C++ in getting robust apps out into the marketplace, switch to Delphi as their secret weapon.
Delphi is predominantly a West Coast and European tool. So, unless you're in those regions, you probably don't see many Delphi jobs adverstised.
BTW, you're so called "Gold Standard of Skill" is a bit elitist. Yes, its nice to know whats going on under the covers (that's what component developers do). But, having a developer who can take the tools I give them and mold them into a production quality app is more important to me (unless I want a component developer).
RD
First, you need to upgrade to D5 (forget D4). The advantages are numerous. D2 was a kludge (like D4) and should never have been released. D3 and D5 are the versions that can be considered "stable" for real development work.
That aside, you do make some valid points.
#1 - Syntax: The verbositiy is inherent of Pascal. But, I don't see how this makes it difficult to read. Good clean Pascal coding makes it possible to write code that is very legible and easy to comprehend. It not the diffulty in reading the syntax that makes C++ programmers uncomfortable - its the lack of terseness and the need to to type more to say the same thing.
#2 - Liguistic flaws. Okay...I'll byte (pun intended) and give you this one. I agree, having to declare everything up front can make life a bit difficult. How many times I wish I had the ability to create a temporary variable in a block of code. To get around this, I use another function or procedure or redesign my code. The optimizer solves my problem while making the code, ultimately more readable (in my book).
#3 - Lack of guaranteed destruction. Delphi will clean up anything it creates on the stack except for variables it creates in the heap. Thus, objects created aren't destroyed because they reside in the heap. A simply try/finally block solves that problem quite handily. But, also destruction of the object that "owns" the object will clean up any contained objects as well. This means that when an application terminates, all objects are automatically freed. The caveat here is that the objects know how to free up any memory they allocate in their destructor or there will be memory leaks.
#4 - Genericity. This is where I do have to concede. Templates don't exist in Delphi at the present time. But, there are are generic container objects and untyped variables (ick!).
Supposedly, Kylix will introduct real macro support - a feature missing from earlier variants of Delphi. I don't know how extensive the macro processing will be...we'll have to wait and see.
One other thing that is missing from Delphi, in general, is multiple inheritence. I don't know if this is a blessing or a hinderance. But, much of this is possible using interfaces which ARE being ported to Kylix.
Yes, I am a Delphi developer. Been so since Delphi 1. Prior to that, I was a C/C++ and VB developer (more C/C++ with VB as "glue"). That was 1994 when Delphi was introduced. And,I don't program in C/C++ any longer (except on Linux...where I prefer Perl for most things). That is because I can develop 24x7 database and internet apps on a Windows platform using Delphi (well..assuming the OS can stay up 24x7).
I find the language and tool well suited for group development in addition to single developer RAD sessions. RAD, in my book, is fine for prototypes...but it generates some nasty looking code that is often hard to follow. Nothing is more important then getting down the concepts and then doing a real design.
The limitations you mention are pretty much specific to the difference between C++ and Delphi. Delphi developers can probably put together a list of things missing from C++ as well. Its all subjective and personal taste. A truly good programmer learns to adapt and overcome. They incorporate the best of what they've learned in other languages and use that knowledge whereever pssible.
RD
I was a bit disappointed upon hearing that it was not open to the general public as well. But, that is what I get for not being a certified Borland tools vendor. Our company hopes to apply for that status in the near future (we've got clients (potential and actual) who have had enough of Microsoft. They want our product and they want it on Linux. Period.).
It's unfortunate that people tend to take these notices out of context and put a negative spin on them. I'm not sure they even belong here without some sort of serious explanation. But, there were people responding to this article who were asking questions like "What is Kylix anyway?". A simple link to community.borland.com would have answered that in a heartbeat.
For those that do participate in the tools program, they are elgible to receive the tools well in advance of the general public. The idea here, as you pointed out, is so that they can port their wares in time for the general release.
Truth be told, Borland is trying to release as much information as possible to help prepare Windows developers for the daunting task of taking on Linux. Linux ain't Win98 (thankfully!). And, it has its own quirks that Windows developers simply can't appreciate until they try it.
Delphi component and tools builders will have a tougher go at it mainly because it is not Windows and doesn't provide the Win32 API. Code that relied on low-level features of Windows simply won't work any longer without redesign.
Application developers won't have to worry about these issues as they use generally don't code to the API level. They are more likely to use the tools created by the component writers (who already dealt with those issuer).
Okay, so I'm digressing a bit. Take the notice for what it is...marketing hype targetted to Delphi developers to get them psyched for the imminient release of the one of the most significant advances for Windows developers.
Go to their newsgroup. They posted new licenses as of yesterday.
News Server: forums.borland.com
I think the news group was: interbase.public.kinobi.
I would have to agree with you on most points. However, it's a shame that you feel like you'll be flamed for saying you'd switch back to Pascal.
As a Delphi developer since Delphi 1 (who USED to code in C++), I wouldn't consider switching back. True, the official PASCAL standard is pretty poor. But, since Borland Pascal has become the defacto standard (at least in the Windoze world), they have extended the language considerably.
These extensions, coupled with the rich IDE and visual component library (VCL) have enabled developers like myself to develop hiqh quality mission critical 24x7 apps (really). Try that with the likes of VB. Face it, VBers, a serious internet/database app probably can't be done that can be run in 24x7 environment.
Why? Delphi provides true exception support, produces true compiled code that doesn't need a runtime environment. And, it can call every Windoze API function in a clear, concise manner.
Even when Windows API calls go South, Delphi can detect and correct most of them. C++ programmers would find themselves at home here.
Database access is extremely rich as well. Out of the box Delphi Enterprise supports, Oracle, DB2, Paradox, DBase, InterBase, Sybase, ADO and ODBC (to name a few). MIDAS and Visibroker make it easy to write distributed apps that can use COM, DCOM, TCP/IP, or CORBA. Extensions to MIDAS make it easy to build XML-based database applications (Internet Express). And, there are thousands of 3rd party add-ons.
Where Delphi gets its bad rep is (in order of highest to lowest precedence):
a) Previously poor corporate management (it's getting better...really).
b) Corporate IT Management that obtains its IT knowledge from rags like PCMagazine (owned, I believe by MS).
c) Developers who jumped on the VB bandwagon because of job availability and salary (see (b)) who then become Microsoft drones.
d) Initial cost of VB was lower than Delphi (now rectified). Newbie developers purchased VB for home and then went looking for jobs (see (c)).
There are some features notably missing in the language (as they are in all languages). I miss operator overloading in particular. Multi-inheritence is something that I wish for on a very rare occurance. But, when push comes to shove and I need to write a commercial quality application that gives me high performance in the shortest amount of time, I'll go with Delphi over C++ (Borland C++Builder included) just about anytime.
The fact that it is now coming to Linux only means more opportunities for developers like myself. I'd love to see Linux on every desktop. My lowly 400mhz workstation boots into Gnome (that'll have to change) in two seconds. I haven't shut my Linux box down in over a month with no ill effects. As long as I can still build console apps and daemons using Kylix, I'll give up GNU too.
Where C++developers/GNU purists get the leg up is that C++ is pretty much an open standard. And, there are good, solid, free tools out there as well to make development possible. But, unless you don't care about performance, executable size, and compile speed, all you gain by staying with open source compilers is just that...free or opened source tools (for now, at least).
If the commercial tools compile the same code (without changes) as your free tools can and you get added benefits from doing so, then why not? Am I missing something to say if the BC++ compiler can compile the kernel and its smaller and faster, is that not a good thing?
RD
Although other OSI project for Delphi exist (most notably, Winshoes (an internet component suit), most of the projects current adopted by JEDI deal with conversion of Microsoft Windows API headers into forms usable by Delphi developers.
As a side note, Kylix (Delphi on Linux) has been demonstrated in Spain (early betas). Supposedly, the demo compiled 45K lnes of X-VCL in 2 seconds.
A port of BCC 5.5 to linux already exists. This was demonstrated at the 1999 Borland/Inprise Conference. Yes, the version released is for Windows. But, they demonstrated that they could run the same source through their Linux compiler, have it compile, and run as expected.
Borland is planning to release Kylix which includes (to my understanding) Delphi on Linux as well as C++ Builder for Linux.
If anyone has used their tools, you will find that they provide all the source code to the libraries (mainly the VCL). They don't include the source to the compilers or other tools (they need to make money somehow, right? Isn't that what for-profit corporations do?).
Nobody is forcing you to download the BCC. Nor, are they forcing you to visit their site. The fact that you went there says you are interested.
The only reasons I can think that they want to know who is downloading their compiler (i.e registration) and filling in the survey are:
1) Knowing if their marketing campaign is having an effect.
2) Being able to notify registered users when Kylix becomes available.
In a previous post, somebody griped about the BCC compiler stating that the Cygnus tools (even MS's own were preferable to BCC). The only downside I ever experienced using BCC was that many libraries in the Windows world simply would not link to it. But, I experienced the same problem with MS's and Symmantec's compilers as well. At least in the Linux world, there is a standard on how these libraries should be built. The Borland compilers should be compatible with other libraries on Linux. No?
So, why does everyone keep knocking Inprise/Borland for their approach of giving away the product?
Personally, it seems to me that too many people don't want to look at the good that can come from this. They expect everything to be free or a threat to their purist, unrealistic, view of the world. That's great for hobbiests. It doesn't make money. Or, perhaps they are staunch MS supporters who don't want to see Inprise/Borland succeed? Just read their posts and decide for yourself.
Your hypothesis is flawed (as was my original statement...I was assuming a sysmmetric cipher)and this would not thwart the detection mechanism. If you are forced to turn over your key AND the key you give is bogus, one of two things will occur:
a) They unencrypted document will contain garbage (I don't think PGP will even let you decrypt a document with a bogus or wrong key). End Result - they nail you for contempt.
b) Any digital signatures on the document will not be verified. End Result - they nail you for contempt.
In 99.9% of public key systems, the document is encrypted first with a symmetric algorithm (like CAST, IDEA, DES, etc) and the key encrypted with the receipients public key.
As the symmetric key is recovered during the decryption phase and the document is decrypted, supplying a bogus key won't help you any.
End Result - Contempt of Court.
RD
Actually, this possible using a technique like Winnowing and Chaffing.
W&C enables more than one message to be stored in an encrypted file. Additionally, garbage bytes are thrown in to confuse the issue even more. Only the intended receiver of a particular message will receive the plain text.
There was really good article on this in DDJ a couple of months ago.
Funny you mention the D.O.S. attacks and Internet filtering software.
It is a sad state of affairs to realize that we, the internet community, are being targeted by laws that we, ourselves, help foster.
Truth be told, cyber-business is here to stay. Laws will be enacted world-wide to curb the activities that a few regengades are engaged in. Billions of dollars were lost this week by companies and stock-holders in the US alone by the fear these individuals have created regarding the security of the internet.
Dot.com businesses are, in many cases, little guys who had an idea and used the internet to capitialize on it. Potentially, many will lose their business because of this so called "cyber-vandalism".
We (the internet community) are attacking our own resulting in restricted access, increased police activity and legislation to deal with these miscreants of cyber-society.
Don't be suprised when measures are put into effect at the backbone level to detect these types of attacks and terminate all access to the internet by individuals and/or ISP who engage in or allow such activity. What about entire countries being denied access? It will happen, eventually. Look at China.
So, who are the victims now?
Furthermore, it can be proven that you committed the crime. All they would need to do is encrypt your "plaintext" version with the key you supplied and compare it with the message they are holding. If the don't match and you don't turn over the real key, they can hold you in contempt at the very least.
In the US, there was a case a few years ago where a suspected child pornographer had encrypted his pictures. The prosecuters wanted him to hand over the decryption keys. He refused on the grounds it violated his rights against self incrimination. Wonder how that one ever turned out....