hotjug

XYZ is dead! Again?

2006-07-14T00:03:00.002-07:00

Isn't the "XYZ is dead!" slogan somehow similar to inventing new buzz-abbreviations like SOA, CRM, PLM etc? Just the two ends of technology lifecycle hype. Considering that there are more than ever open source or free development tools, frameworks and code snippets available for Java and most probably, because of fearing loosing the momentum, Sun is making Java open source, I would predict the next headline to be "The death of J2EE is greatly exaggerated" Yes, Java platform has had some genetic defects from the very beginning: speed/performance, steep learning curve, lack of readily available developing/debugging tools, lack or not very good implementation of some concepts of OO. But the level of sophistication it has reached, is just a sign of a non-fatal middle-life crisis. The best that can be done is to go back to drawing boards and re-invent the Java platform, but does Sun have the will and resources to it. The VB guys are quite in a situation - due to cut of support by Microsoft, the only choice is VB.Net and Visual Studio, .NET with C# being called by some people _the_ Java 3. Who is afraid of platform/provider fix, can ease its fears with Mono - the open source alternative to MS branded common language runtime. But be warned - .NET framework 2.0 has already been a confusing mess of new features and annoying defects. The 3.0 brand is going to be even more confusing by embedding many of the Vista elements in it. The PHP folks are also on the edge of plunging into the sophistication limbo. Limitations of the natively procedural language have forced PHP language developers to embed more OO into the language and there is a myriad of PHP frameworks in development, most notably Symfony and CakePHP. HTML will remain separate and whateverML for the forseeable future, but definitely some promising new 'platforms' like Ruby will pop up. It remains to see shall they become the new de facto standard for ad-hoc or enterprise development. The story of Python - a language quite similar to Ruby - that was a hit 6-7 years ago is currently a niche player and it seems it will remain to be so. Surprisingly, some of the old school languages like Perl and C++ are well and alive, have their own niches Perl for administration and utilities, C++ (universal but) for platform independent GUI development. The next version of Perl is promised to have even better OO and the popular C++ frameworks like Qt are steadily developing their feature set. Overall - is there any language + framework + development toolset that we are really satisfied and with no signs of deterioration? I doubt. There are trends and not that steep to either direction. However the next bold 'perfect' language and platform should definitely aim to be the following: - open source - platform independent - fast - highly OO - support or embedd top class developing and debugging IDE - easy to get started, good documentation and huge community - good for both web and fat-client GUI-s The James Bond of the platforms/programming languages? If you spot one, drop me a note.

Nokia 770 - a product strategy thinking excersise

2006-05-12T15:21:00.000-07:00

Besides the nice feeling one has when reading about, seeing, playing with a new gadget, N770 makes me feel like it does have lots of strategic thinking put into it. And yes, it provokes some constructive scepticism. So far the PDA/cellphone and Tablet PC/Webpad form factors have been most widespread. The 'in-between' sizes have been there (oldest non-rugged/industrial one that I know of is Casio FIVA) but none has had any extended lifespan. So Nokia came in with just a couple or right decisions:

- price
- form factor (you may want to stretch it a little bit)
- open source from ground up (that will probably solve the lack of applications in the meantime)
and has the problems of
- performance (maybe the software on it is more resource hungry than it should be)
- battery life (improvements in power management?, alternative processors?, 'flash HDD' versions, but certainly a 3000 mAh optional power-brick+external charger)
- a little bit exotic development platform. (Don't ask how, but you have to make it work on Windows, including an emulator and support of some decent IDE)

N770 does not and will never make a good gaming machine.

- It should do good browsing (some stripped down version of Firefox),
- have all possible utility apps (PIMs, e-mail, VOIP, light-weight office apps + drawing-sketching - GIMP-like or 'SketchUp embedded' etc) and
- stress connectivity (I would add Irda and software for home electronics control and co-market the device to home control companies like X10 and others).
- Have an out of the box N770+Windows PC optimized Wireless Display implementation (forget Mira)
- Hopefully, the acquisition of Intellisync also brings in seemless integration (free availability on Maemo?) with all mobiles and desktops (Outlook and Thunderbird)
- N770 can benefit the most from the current Ajax fever. Set up and try a full set-up of Office 2.0 apps and put some of those companies into Nokias shopping list or provide or co-provide with, say Google (wink-wink) some of the essential services like drivespace, mail etc
- Another area is in-car/GPS use and possible more serious relationship with the likes of TomTom.
- You may consider looking at options to increase movie watching performance (specialized chips) without too much increase in cost

Peripherals need more attention and I guess - from Nokia itself. Namely

- mobile keyboard input (buy/license Twiddler2 and make it finally Bluetooth, iFrog or the new IR virtual keyboards, Pepperpad type keyboard is worth considering, maybe as a plug-on) and
- a set of stands/docking stations/car mount kits. The 'stationary' part of mobility should be there without the hassle of building it yourself or hunting for exotic gear.

Yes, UMPC, handtops and downscaling TabletPC-s (Motion LS800) will be there for a while, but no-one beats the price and the Nokia brand. Hopefully it is a sufficient laundry-list to make N770 the coolest device on Earth.

Digital Fortress, TRANSLTR and ScreenLock

2006-04-23T11:38:00.000-07:00

Somehow I got into reading Dan Brown's DF. I have not been much into reading during the last years, but I really gave it a try. In spite of all the prejudice due to the publicity and popularity. I read half of the book and started to feel bored. Too many errors and nonsense, however, as a storyline it worked quite OK. Now, after having read some of the more critical opinions on the web that I would not like to repeat here, I had some random thoughts:

DF was suppposedly his first book, so he just designed and tested the style. Ok, and not bad for a first book. I would be happy with a first book like this.
As with media, commercial book publishing is not up to factual correctness or high artistic value - what makes money (interesting and easy reading that allows to build a buzz of w.o.m, PR and viral marketing) is good enough to be published
Similarily to the davincicode, if someone is looking for some message behind the factual errors and illogic, claiming that there is some überlogic or message behind it, well... cool down, it is just some laziness on the editorial work, because not having an army of researches means that the book is cheaper to produce
The level of computer knowledge represented and unfortunately also some logic represented, well describes the overall understanding of the computer user masses. Helplessless with anything more sophisticated than starting a word processor, it well is magic and mystery
DF is the soonest to be obsolete book of Dan Brown. Computing develops so fast that when today it is a joke, tommorrow it is bloat.
The security paranoia it possibly provokes in an average computer user may be even positive. However, it covers only a fraction of issues an average computer user may face. So the result is more like having a frightened to death and prompt user on danbrownish issues, acting like a suicidal blonde on a myriad of others.

All in all, if one looks some generalization, the starting point of which is this book, I would say that the book plus the reception and interpretation is, not so much art, but a journalistic excersise or social experiment that describes well our current society. And that, well, is good art, if you please. Next time if I am after some literary experience, I'll pick up some literature. Promise.

Undercover Intel Op-s and Natural Development

2006-03-13T00:28:00.000-08:00

The history is full of conspiracies, secret killings but nowhere at the perceived amount and level of sophistication as during the 2nd half of the 20th century. It was a policy of US intelligence until banned. Remember Castro and a number of Latin American and African suspected intel killings. If we were the good guys, then as a substitute of a military invasion it looks reasonable - surgically remove the root cause - less costly both in dollar terms and human lives. Assuming, that as we act as Gods we do have the level of morality as the God. So we have several Gods. At least the intelligence agencies of UK, France, Israel and Russia are known to have had more than one targeted killing in their history. From a purely ethical point of view - we do not have this right to kill. From a political-military pragmatic point of view - it is a cost effective way to supposedly influence the situation to the desired direction. The hypothesis is - secret assassinations is such an unnatural way of intervening that it probably leads to undesired results. An enivronment that has allowed a despot/manipulator/leader become to power is most probably best kept in balance by the same d/m/l. That doesn't mean that you should do nothing. It only means that there are other ways. Maybe simpler, maybe much more difficult. Just one recent addition to a list: there was a plan by British intelligence to assassinate Hitler. It was dropped. Partly due to the impossibility to complete it at later stages of war, but the most significant other reason seems to be that the German Army leadership and politicians would have been released from the sick, narrow-minded approaches of Hitler and that would have possibly lead to the containment of the war. Drawing back, closing fronts, recovery, status quo.

Dichotomies and Solutions

2005-12-12T12:41:00.000-08:00

Many if not most problems in this world have two controversing solutions. You have fundamental issues like 'does god exist' or is 'ford better than opel'. Taste and personal freedom apologists leave it up to the individuals and it gets rather dull if not un-social. Pure rationalists always try to find a syllogistic proof for things. The one thing we are also not interested in here is propaganda, brainwashing and belief/praising because of fear, dependence et al. Once, I was a big fan of logical reasoning - until I studied a textbook of logic. It proved a few important things a) the concept of syllogistic logic is flawed and contradictory for sophisticated and real-life problems (the most fundamental issue being that all logical systems, at some point, go into contradiction with itself) b) it is usable to prove some particular point after a creative process, but it does not help you much during the creative process itself. If we step a little bit out of the syllogistic logic, mix and match what the american school lawyers do in reasoning you can basically prove whatever point you need to. A perfect endless story is the existance of god. OK, you all know that between the yes and no there is a variety of incarnactions of god being, half-being or super-being. Also, it may be irrelevant or the answer to the existance of god is 'my teeth need brushing'. Trying to find a common denominator for all these solutions is quite hopeless, but nonetheless one of the options. The answer to the existance of god may be the act of passionately or not searching the answer. It is not only conflicting theories it is also about meta-understandings that do not fit to the dualistic or logical undestanding of the world. Take the theory of relativity after newton, then some additions from electrodynamics, quantum mechanics and alas - no commonality with gravity. The fact that the unified theory is not there may be good reasons that it will never be, unless we morph again our understanding of what and how we are looking for. What is it? Mind twisting? Some 'better keep off from it and live a peaceful life' thingie? For me this is a chain of reasoning that is elaborated here less that necessary and one of the most essential principles I have discovered and what I try to use in everyday life.

Change - Inevitability, Result, Purpose

2005-12-12T12:03:00.000-08:00

Things happen to change or more precisely people tend to change things. Lets take only the social side of things, for simplicity and concentrate on these items: organizational structures, politics and product development. Change happens usually, because something is wrong, bad, not as supposed to be. Then people change things - put/move right people to right places, subordinate this to that and fire some. Same thing in politics - things are always less than ideal it always gets better with some change. I am not talking of the institutionalized 'change' brand - just only about the description of the results. Are things really so bad or in need of change in all these three areas so that this amount of change is inevitably needed that we see these days? Organizations - for the critical observers there obviously are wrongdoings and excess trembling in changing the orgcharts, in politics - preaching change is the only way to increase (or not to loose) your poll share, in product development - the only way how to keep people buying is to add that watermelon flavour to your chewing gum and that extra n-th blade to you shaving gear + turbo + laser guidance. Suppose, from the objective point of view, no change is needed. What happens? You can be sure, some will start to talk about 'nothing happening', usually the same ones who whined about 'too much change'. Politics - a less controllable environment - it is is not possible there. Whenever the big guys cool down on change, some newcomer finds a way to change-talks. In product development/marketing it may be fatal. No change, probably means going out of business. So, it does not matter, if you need much change or little or is the need for it big or of a lesser kind - looking back - the result is some sort of change. Thinking of the passion before and during the change - what a big work it was to engineer it and what a minor move in the long term perspective. If you have a defined reason for change, it often binds you to a problem to solve and that may constrain your creativity. Take a third point of view - without knowing about the details of the objective need for change would there be any difference between the two options: a) real need for change and a change b) a faked change to keep things bubbling, but reasonable This leads to a new paradigm of change: a) whatever it takes, you have to introduce some change b) the result of a change should not be stupid, but also not overly beneficial c) predictable change is not change, 'much' change is radical and overall not positive, what you need is 'different' and not too much different. And of course - with small incremental changes, you can teach a giraffe to fly d) the setting of a problem as 'change without no rational reason' leads you to the undiscovered jungle of creativity, the vanguard solutions may actually show you problems that you have not discovered yet or give you a pro-active edge for future problems. Change is a bad solution to problems. Change for the sake of change is bad. Change for the fun of change is good change.

The Last Reason Why e-voting Will Not Work

2005-12-04T10:58:00.000-08:00

There are already tens of arguments against e-voting or e-ballots. The ones who do not know cryptography and technology feel insecure, I don't. There is always a way to set up Anns and Bobs so that everything from the press of a yesno button till summary statistics is true and objective. The main and absolute argument for e-voting is of course, convenience. For some it may even be a more objective way to get the results. But I think, that after numerous experiments, the classic 'go to the voting room and vote' model will prevail. Surprisingly it may end up so that the actual voting is not done with pen and paper but with a voting machine (hopefully not with the hackable ones used in US), but the important part of voting is 'going to somewhere special and voting there'. Actually, no voter identification technology is embedded into the voting terminal, it is done before and is independent of voting. There will be other bits and pieces to this, but generally the above describes it. Now, finally, the last reason(s):

Ultimately, however good are the identification technologies, it is not possible to control, is the person voting actually the person voting;
Ultimately, it is not possible to control, how and by what, a person at the very moment of voting is influenced. Consider going to voting as something similar as going to church or to a doctor;

As voting is about control and influence 'in a situation', this makes a huge difference.

Ultimately, technologically, there is a place and time where the voting result and voter identification information co-exist and there is a higher than 0,01% possibility that the voting will not be anonymous. Ok, it is the least important and probable, but still the 'ultimately' reason.

Everyone is talking about the technology and cryptographic background of e-voting. I think, this is a moving target and is improved over time. AND, this is not the issue about e-voting. What IS important is the social and behavioural models around e-voting. And in this area, research is seriously lacking. And really finally, a hypothesis for thinking: Suppose we have two identical countries. Absolutely identical. Even the flap of the wings of a butterfly is identical in these two countries not to talk about the long-term political preferences of the populations. Suppose one of them is voting the classical way and the other is voting the e-way. What is the difference in the voting results? Lets eliminate also higher participation rates due to accessibility and outcome differences due to that. Lets assume that they prove to be completely honest and transparent ballots as much as it depends on the overall levels of democracy in the twin-countries. Is there a residual difference that is caused only by the different technology used?

Technology changes the perception of consciousness

2005-12-04T09:58:00.000-08:00

I don't recall exactly, how old I was, when fantasizing about different computer user-interface technologies, but I finally got to the point of connecting the user-interface cables directly to nerves/brain. It was not an excersise in Sci-Fi (what I am not a great fan of) , but a consequential thinking project - how to get an even better visual, audible etc interface to computers or video and audio devices. At some point I had collected lots of information about experimental holographic visual reproduction systems and near-reality audio reproduction systems consisting of hundreds of loudspeakers. No problem with these. It got more difficult with senses like smell, taste, temperature, tactile, cinestetic and all the rest. It was not a big problem regarding voice recognition and basic machine vision but the difficult part again was to provide full body motion input. So, not Sci-Fi but a result of a long thinking process fed by technical and popular science magazines. Now think of the consequences. Suppose we have the connectors on our body or it is even wireless and we have the device that emulates the reality. There are myriads of developers for that virtual environment. Where does it lead to? Yes, correct - a big part of us do not need reality anymore or reality just becomes a very elitarian entertainment, maybe even too costly to afford. It takes time or some very good reasons to reach the point where the real-real world as we know is just a huge repository of interconnected VR computers and brains floating in tissue culture media. Is this just wild fantasy or a thing quietly happening under our own eyes? I guess, yes. The foremost area of development is game technology or the virtual environment where all this will start to happen. Give it another 10 years to develop and we have a place to be in. The user-interface technologies is a little bit bigger question mark. What is the most crucial point is actually the social/economic/political reasons to start mass-adopting this technology and say, erecting VR brain repositories. Some people, like game addicts would welcome readily this brave new world. Of course, there will be human rights and health activists etc. The more important part is, what could be the practical reasons (may be not as brain repositories, but at least with connectors on our necks):

Good entertainment, amongst other ways of relaxing;
A more feasible / efficient working environment for some professions. Reality is not just the right place for some jobs;
Addicts;
A way to gain control over the minds of a bunch of people (and how to protect against this power);
A neat way to solve population/urbanization/consumption problems

But I think, the most important reason is not in this list. What is it? This technological possibility, actually gave part of an answer to my philosophical wanderings during the high-school. One of the big questions in philosophy is "What is reality?". Actually, the mere possiblity that we already are just a repository of neatly interconnected brains and VR computers and there is no way to prove the contrary, puts the whole classic discussion about reality under question. The Fine Print: Later I discovered that the Russian Sci-Fi writers Brothers Strugatsky had this concept in one of their novels. Also, lately I discovered that the movie Matrix is based on this concept. Finally: Read once again the heading of this post.

The future of ERP

2005-12-04T08:25:00.000-08:00

In this world where vendor consolidation is still the most significant thing to happen in the business applications business (ERP, CRM, BI, EAI, SCM etc), the one thing that keeps me wondering, is that what is the medium term (3-7 years) future and trends of this business. This is the list of observations from the actual world (positivistic approach):

OK, consolidation. If you have not noticed between the lines of the consolidator press releases, the ultimate purpose is to reduce the number of solutions, because it is more efficient. No matter that for business (revenue) continuity reasons, some companies support ancient products and technologies. Yes, there are the aggregator, integrator and independent business unit strategies, but unification is still the most viable strategy;
Platform shift. The most significant development stream is technology. Either it is going .NET or Java or advancing its own developing framework. You can be sure that no-one is standing still in this agenda item. It is not the lack of strategy on this, but overly ambitious or obviously wrong choices should keep you alert. You may put also change of database platform as part of the platform shift. So if someone running on file-server databases and is ignoring or not paying enough attention to its up-scaling customer base, you should change the vendor;
Functional enhancements. Actually, this is the only thing that the users want. Even if the technology is ancient, if the product is functional enough, there is no reason for change. The devil is in the details. If it is bugfixes, is OK, but if it is bugfixes only or defects converted to enhancments under the name of functional enhancements only, you should be careful. How many truely new functionalities your supplier has introduced lately? Usability enhancements do not count.
Interoperability - YES, all that e-everything, SOA, mobile, SOAP, XML rubbish - NO. Do not confuse this with technological advances or functional enhancements. All too many suppliers do these things only to have that one extra press release to contain all the buzzwords that all the others have. It is mostly about the possibility to integrate (custom or standard) and enable access (www, web services) to the system. As a sub-topic here - everyone who works to enable thrid party development using non-native platforms for the solution, is on the right track.
Open Source Applications - I would love to announce that there is/are open source applications for the key application areas for todays business needs, but unfortunately the answer is still no. Compiere, SugarCRM, BIE, JasperReports etc have a long way to go both on product maturity and community acceptance before you can take them really serious, but it is a totally different world today already compared to the time 5 years ago, so thumbs up!
Source Sharing and 3rd Party Development - Actually, what needs to be paid attention to, is the keen closed-source'ers. If it is SAP, I do not care. If it is other than that, they must have a really good reason besides the intellectual property bla-bla. What you loose is consumer confidence, 3rd party developer community, a bug fixing parachute for VAR-s and ultimately, the future. Surprisingly, even the business model of MBS is based on the availability of the application logic related full source code (but not the proprietary runtime and development platforms Dexterity, MorphX, C/SIDE) to the VAR-s and if necessary, to the customers.
Acknowledging horizontalization and verticalization - Simple examples: the one who thinks that it is worth maintaing its own proprietary report designer/generator, BI app or start a from-scratch development of a CRM is wrong. It is a waste of development money. Also, the one who does not satisfy the temptation of the end users to get everything conveniently to and from Excel, Word, Outlook or equivalent widespread apps, is in a process of loosing to the competition.

This is the list of would be scenarios, what could happen, what is thought to be the futue:

e-commerce. Being first, does not mean that it is the most significant. An old thing that everyone has or will have or develop in some form. After a period of e-pessimism, developers are again introducing integrated e-shops and web-based interfaces to the data in their systems for end-users, customers or vendors. The technology is now cheaper, internet more widespread and solution knowledge more readily available. I would not qualify it as part of horizontalization.
100% web based apps. Extreme optimism and pessimism have accompanied this technology paradigm. It is worth being less passionate about this trend, monitor and adopt what is feasible and appropriate. Web based apps are quietly gaining presence (crm-s like salesforce etc and several commercial accounting apps). The Ajax / Web 2.0 hype may make it even more widespread, but things like decreasing cost of broadband and wireless access, decreasing suspicions of end-users regarding speed, security and fault tolerance of such apps are strong trends that determine part of the future to come. The trick here may be that the fat client guys need a web-thin one an vice versa. Maintaining a good app on both platforms is very costly, so the one who has the best solution to this question is in the winners league.
True and usable implementation of the interoperability buzzword technologies (XML, SOAP, web-services, SOA). XML - if it is using some open standard shemas for data exchange that is useful - it makes sense. If it is proprietary and only adds to the sophistication and 'slowness' of the already complicated app - it is a warning sign. A perfectly wrong application of technology would be to port all the functionality to web services assuming that users need it and it is just technology-wise fancy. Wrong again, even Microsoft suggests using RPC calls instead of SOAP and web services for connecting back-ends with UI.
Granularization of applications and functionality. This would be a possible scenario, but it has its flaws. The idea is that developers do 2 things: a) they enable interoperability of their existing systems with other systems using some widely accepted standards and technologies and b) new and existing developers make their applications compatible with these interoperability standards. The wildest dream in this realm is a best of breed Accounting General Ledger that you can use with say, a best of breed Account Receivable package with no or little installation, setup and data exchange latency. The less wild dream would be to have ERP/accounting packages of some providers seemlessly working together with CRM or e-commerce systems with some others withouth the painful and costly gizmo called and EAI (Enterprise Application Integration) application or yet another proprietary bespoke integration.
The share of subscription / maintenance / service based business models is growing. This does not need any serious argumentation. Most of the popular web-based apps are subscription based. Look at salesforce or netledger and many other similar apps. The old school ERP companies live out of maintenance revenue, so they will also consider pure subscription based models as competition may require it. Microsoft is bending the fine line between subscription and one-shot purchasing even more having everything in between and models like rental, full solution financing etc. Also, do not forget the open source models of providing software for free and providing all related services for a cost.

The conclusion of this post is actually "To be continued...". I just cannot recall all of the trends/scenarios that have passed my mind.

The problem with (advertising financed) media

2005-12-04T07:59:00.000-08:00

9x.99 % of media (newspapers, magazines, TV, radio, web) is financed by advertising. It is all so obvious, but we tend to forget it in everyday life. Simply, subscription-only financed media models have proven not to work (unless it is some attractive premium content like pay-tv or it will be required by law ;) ). There are two problems to think about here: a) some idealists may think that the purpose of media is to 'provide comprehensive and objective information with the shortes possible time delay', but they are wrong. The purpose of media is to 'attract the maximum potential media consumers to their channel, so to get the largest / best etc audience to be presented to the would-be advertisers who will finance this media channel' by paying for the ads. The result is that instead of information according to the first definition, you get attractive and biased bloat; b) The second problem we have is, that the advertisers invisibly influence the content of the media. Unless you keep your mouth shut or you write only the good news, we draw off our advertising money. So, the next time you are surprised, why some newspapers do write about some topics or bend some information in a strange way, you have been warned. As always, no reason to be pessimistic about it. Just keep it in mind. You can always comment that the above is not true, but I am more after the solution than the pessimistic halleluujah about these posts.

Why the ultimate e-commerce model - comparative shopping is seriously flawed

2005-12-04T07:30:00.000-08:00

Pushing everyting 'e' to the max - skipping disparate single vendor web shops etc aside - the ultimate salesplace is a comparative shopping tool - information from different sources is aggregated to one single source of comparative information. So, when you go there and you need a hoover, you just punch in your criteria and you get the best offer. The 'best' often remains to be defined by you. But do we have many of these comparision shopping sites with considerable commercial success? The answer is no, with a few notable exceptions like computershopper.com and some consumer society sites. The reason? Is simple - this business model does not work, for these simple reasons: a) we may think that the 'perfect market' defined in microeconomics as - having uniform (comparable) goods on the supply side - equal amount of information on the demand side - no supplier or buyer can significantly influence price buy supply/purchasing dominance - transaction cost do not significantly influece (de-equalize) the opporunities of the parties This perfect market is simply NOT something that the suppliers want; b) the suppliers are well enough in a situation of market imperfections where they have some sort of direct access to the consumer and by having that they can make their sale. c) finally, whatever the qualitative criteria, mass-consumers tend to decide based on price, so the loosers are premium pricers with some extra bells and whistles even maybe at premium quality, the winners are cheap(er) and less that desiarable quality suppliers; d) it is simple to compare things based on price. Cash is a good uniform measure. At the same time, it is difficult if not impossible to present qualitative data in such a form so that it will satisfy the variety of would-be consumers in their comparative shopping; e) the actual cost of providing the comparative research is rather costly. It has to be financed, but the classic advertising model bends the consumer decision making towards the ones who have larger ads in premium places, thereby distracting the suppliers that do not afford advertising f) and there still remains the question, how objective is the research. If there are many of these sites, ok, but what is the guarantee that the reviews and comparisons are impartial? Is there a solution? Possibly yes, but there may be more than described here. The technological part can be solved so that the cost and sophistication of aggregating the data is reduced to minimum by aggreeing on publishing product data in a uniform format (agreed XML schemas), so the cost of maintaining the site is reduced to minimum. This is an issue of standardization and the pros and cons remind very much those discussed in here. Also the way reviews are compiled and published can be standardized so, that it will keep things objective enough for the purposes.

Open Source, Closed Source and the Not So Open etc Source

2005-12-01T10:03:00.000-08:00

The ones who know the benefits of open source, do know and adore it. The ones who are the apologists of closed source, do what they know. So it is not about this eternal Linux-Mac-PC war. The problem is, that even if a project is defined as open source or similar (GPL, LGPL and a myriad of other varieties) the benefits for a would-be adopter are strikingly different. It is different for the developers, for the managers/decision makers and to the overall interested community. So, what makes the differenece (in order of my perceived importance): a) main platform - it is considerably more problematic to have something written in Java, Phyton versus PHP, Perl or Visual Basic (yes the old one) as the development communities and the number of affordable but still skilled developers differs hugely. It is not that one is better than the other, but about the adoptability, affordability and future proofness of the platform; b) the quality of the code already there and the amount of native documentation to maintain/develop the solution further. Obvious, but too often the only thing available of a fancy app is the source and maybe the executables; c) presence and size of the development community. Obvious again, so lets do not discuss it; d) the use of commercial components and developer tools. Researching the developer documentation of some supposedly FOSS apps reveals that in order to do full fledged development you need some commercial component, like say in case of OpenWorkbench of many of the Qt based apps (correct me if I am wrong). Of course, you may consider to pay a candy bar price for a three course meal. What a deal! e) existance of commercial support. I would put it as the last one here as say, support of PostgreSQL is a nice thing but Oracle buying InnoDB creates mixed feelings. So it is not clear where it leads to, but definitely it is beneficial. All too clear to put in a bulleted list, but overall the success is also determined by the maturity, stability and feature set of the would be app, either at launch on at the current point of time. So, the conclusion is: open source may cost you something and probably will and the key benefits may come from other sources. The problem is, does the comparative value of the costs and benefits leave you with a net red or black.