Yeah, I was prototyping this in Milano airports as well, around 2006. Using RFIDs instead of barcodes on luggage tags makes a lot of sense, as you can store within the chip itself the complete story of its journey, instead of having to store it in airport databases, and later retrieve it wherever the bag is. The dream scenario was to store your hotel address on the bag and have it delivered for you all the way there. Also, RFID readers can scan a few hundred chips or more at a time, so it is easy to install them over doors hand have them check automatically everything that goes through.
The project was a great success from a technology standpoint, but it did not take off because: - it only works good if many airports adopt RFID, not just one - RFID chips at the time cost 1 eur apiece, much more than printed barcodes (and there are *a lot* of bags travelling on any single day)
Nice to see that things are catching up 10 years later
Disclaimer: I've been working in the CMS area for the last 8 years, not KMS, but here you go:
what you are looking for is a project, not a product
defining taxonomies, tagging content, making people follow agreed-upon workflows, naming conventions, 'this goes there' conventions takes a lot of time and effort
the best "tool" you can buy is a dedicated knowledge management official (the 'webmaster' of lore;-)
2nd best tool is buy-in from all parties. KMO can train them, provide them with self-learning material, reassure them, etc. Better convince top-level mgmt first, though
KMS products are just toolboxes. Each take a lot of config time, learning, setup, workarounds
go OSS, as you will need to tweak and patch them as needed. Avoid proprietary like the plague
pick something running on SOLR - it has a diverse ecosystem of semantic analysis plugins, and its world-class in searching
you might want to start using a Document Management System (Alfresco) or a CMS with a flexible content model (eZPublish)
Confluence is just a glorified wiki, it leads to duplicate content everywhere because of its simplistic permission model, document model and missing multi-positioning in the content tree
I see lots of comments trying to shift the blame from the administration that originally dumped the chemicals in what was at the time countryside. Funnily not a lot of them highlighting the fact that dumping toxic waste, wherever it might be, is just plain BAD. And this is a textbook case: you might never know what will happen there in 50 years, let alone 500. Food for thought for nuclear waste dumps...
Was it the american indians that encouraged to think about the consequences of one's actions for seven generations?
Totally agree - and yet naming the eZ Components that way lost 50% marketshare to the zend framework on day one. Talk about fads not existing in the it world...
100% agree.
Install filemon,diskmon,regmon,portmon and off you go.
Install Process explorer as replacement for the rask manager.
Read MR's blog to learn how the big boys go at it.
Look for hints at the event viewer: any suspicious/recurring alert in there?
Little hands-on experience with eee 901 xandros and wife: - eee is very well set, up, hard to tell it is linux, and the gui is 100% locked up: good for average joe, but wife wanted to have thunderbird on 1st menu page, not 3rd - and off we go with wikis and the usual linux do-it-yoursel approach - wife was very happy with games: more than windows, and coming witha tutorial, too! - firefox, skype: no problem in there after I set up the accounts (easy: she used the same sw on win) - ooo: complete failure. Being a math teacher, she relies heavily on inserting formulas in docs. The interface in writer might be a bazillion times better than in word, she still did not want to learn it. And all the options to disable automatic reformatting of text are not in the same menus, so disabling them also proved to be quite hard.
In short: the devil is in the details - every average user has a few specialized needs and expectations. Getting everything right - and making it as similar to windows as possible at the same time is an impossible task
That was about the first question that came to my mind: are the Python/Parrot/PHP/Ruby guys aware of these developments in js technology? Are they already porting the techniques? Or is it impossible to take advantage of them because of the basic assumptions in the underlying VMs? After all, despite all these languages interpreters being mature enough (erm, maybe the P ones...), javascript engines are seeing a tremendous advance in speed after many years of relative stagnation...
The grand-unified-find-interrelations-between-your-sql-and-your-code was the most over hyped useless garbage I have met in a while. It eg. found in delphi code that a button named the same as a db column was somehow tied to the db. Or in php code, it was not even able to tell what KIND of databases the code would connect to, since db connections depended on ini files, and it could not grok includes based on variables.
In Italy, newspapers do the "omfg! we are running out of engineers! 50.000 indians or chinese will need to move to italy to compensate shortage!" article at least every couple of years.
The fact is, there are a lot of companies that want 50.000 engineers paid 800 EUR a month. And a lot of engineers that have a very hard time finding a job that rewards them 2.000 euros.
Nuff said (but the uncoolness of studying math or physics is biting hard universities, and in the engineering degrees all the pupils only go for the management-oriented curricula. This is going to prove a problem in a couple of years if not further away)
way to go, bofh spirit. And what if the client is a better dba than you and has a very precise understanding of roles, grants and privilege separations in Oracle?
Your description really looks like the plot for the Ranxerox comic - where a few years in the future this huge, exceedingly brutal, dope addict robot is called in to hollywood to play Fred Astaire in what is planned to be the biggest remake of all time...
BTW hey, nobody ever made a Ranxerox comic! Studios, do you hear me? Here's the next movie idea served to you on a plate!
Sorry to bust your bubble, but shared bikes are all the rage in France right now. Lyon has them, Paris too, and all other biggish cities are scrambling to get theirs.
All it takes is an almost-civilized nation and some careful execution.
The shared bikes-done-right involves: - you only ride them for free for 1/2 an hour, afterwards it's paying - this means you have alot of incentive to never go with the bike too far away from a parking lot - about a hundred thousand parking lots covering the complete city - at every parking lot a pc tower has a map with the nearest other parking lots - bikes can only be rented after you sign up for your rental plan, at no cost but involving a caution and a bank account - bike parking lots are web connected, you can find their placement on the web and even reserve one bike for when you need it
In paris it has been going well, and the city life is though, by all accounts. Bikes stolen/broken/vandalized have been in the range expected by the programme (ie there is some cash allotted from day one for repairs / refurbishing)
Everybody knows how the story goes: after mighty mouse experiments, both Russia ans the states turn to secret experiments to build human superheroes for the army, and it ends up in WW3, the world being destroyed and a new kind of humanity in search of a new planet: The One
Yes, there indeed are laws regulating the 'printed' press, and it is the main justification for extending the law: why should slander/libel be condoned on a highly visible blog more than it is on a newspaper selling a few hundred copies?
Also note that Italy has a dumb anti-terrorism law that basically forbids everybody but the telcos to provide free wifi access (people who do MUST take a copy of the id card of every wifi user that connects to their hotspot, and keep logs of all web access), and it is really hurting the mobile workforce.
US is not the only government with police-state temptations that needs to be kept in check...
Not bank related, and not in the US, but this happened to me: - telco cut my phone cable, without any previous warning, because of what amounted to an accounting error on their part - I had to jump through loops for a week before telco admitted their fault. Initially they blamed it on competing companies requesting to switch my number over, and competing telcos denied that. Note that all of the telcos only have customer desks that respond via call-center: not very easy to call when you've got no phone (tip: calling call centers via mobile costs lottsa money) - after admitting their error (even though the disconnect request in fact did came from a competitor), they gave me the connection back - within a month... -...and they charged me 150eur for the favor (the price of a new connection)! They stated it was my duty to recover money from 'the other company' - I refused to accept that logic (THEIR FAULT, I HAVE TO SUFFER), and decided to switch operator... -...but all other operators told me that they could not provide me a contract using the same telephone number, because the original company would never surrender it, since I had some debt pending (the fine I refused to pay). Note that they had already surrendered the number in the first place, breaching in fact their contract with me, but they would refuse to do it when asked.
This long post just to clarify one thing: you do not always have the choice to change banks/telephone/services providers. And the logic 'customer always at fault' does indeed apply 95% of the times
Not all airport display systems are created equal, but to use a full fledged PC with a MS os onboard is not the only solution. For biggish deployments, every monitor usually comes with its own single-motherboard-industrial-pc in the casing, possibly running some sort of embedded linux distro and either running a custom app that receives raw data from the central server, formats it and displays it locally or just running an xserver to drive the display with the application running on said server. Similar setups (citrix/terminal services on wince) are possible in windoze land. In such cases, there is no restriction on cabling, since transport is standard ethernet, and very little cost saving...
I guess different people have different learning approaches, so what is perceived as "easy" or natural by someone is deemed obscure and arcane by someone else. In general, I prefer to read the 'technical specification' manual rather than the 'filled with examples' tutorial, but I understand that in order to get any benefit out of it you need at least a basic understanding of the topic at hand.
More to the point:
- I find Apache docs very good, both when compared to many other OSS projects and when evaluated on their own. The basic params are described one by one, and you can usually find the correct combination with a little trial and error, but for the most complex tasks there are dedicated sections which are very helpful (eg tuning)
- 'using a text editor' to edit the config file has a very nice side consequence: it makes backing up the config and reimporting it later / duplicating it extremely easy. But I am sure a lot of other/. readers have duly pointed out all the many advantages that apache has (such as using a simple search command to find out if a parameter exists/has been set: it can be a real time saver compared to the infinite series of tabs/dialogs that have to be navigated with IIS when you are not really sure where things are set)
- by your own admission, you have never configured a moderately complex website. I would consider the fact that you do not know what http compression is a sore spot in your CV, but that is just my own personal opinion. IMHO the good coders are the ones that have a passion for learning (at least at the basic level) all the aspects involved in their work, be it system administration, network protocols or memory allocation / cpu architecture.
- to be honest, IIS is indeed more friendly 'out of the box' to the complete newbie. It complements that with the sheer inability to implement all of the complex setups that apache can do, and with an exponential growth of the time needed to configure it vs. configuration complexity (the example about setting folder perms being a prime example: it's fine and dandy as long as you develop your small app over which you have complete control, but just imagine having to manage tens of different web applications developed by different people, that do not write down clear specs for configuration and have very sloppy security practices, and you see the mess it becomes. Last week I had to deploy an app. where the file containing the db credentials - in clear was set to be world-readable... I was NOT pleased). Otoh Apache has a higher barrier to entry, but it gets easier to use the more you know it. This is a well known paradigm for both OSS (makes easy the hard things, but not easier the easy ones) and microsoft sw, which makes extremely easy the easy things and impossible the complex ones.
Ok, so did you also try to: - read the apache manual that comes with the distro instead of googling around as 1st stop help shop - add to logfiles a custom http request header (requires writing a custom dll in IIS) - have different logs generated by different virtual hosts / folders / requests - edit by using a text editor the iis config / back it up or restore it / write a plugin to manipulate it (impossible up to iis 6) - use ldap as authentication backend for IIS - find out the access perms set on a huge tree of IIS folders without clicking them open one by one - set up decent http compression in IIS without resorting to external tools such as Port80 - set up IIS as reverse proxy or load balancing proxy
sorry, but as far as ease of use, configurability and sheer number of features there really is no match. Apache can be tweaked to serve every possible imaginable unconceivable convoluted application you design. IIS cannot.
Lots of people nowadays will be happy to tell you how there are 100 (free) MVC php frameworks and templating systems out there that allow "proper" enterprise web app coding instead of the ugly db-logic-presentation mixup that php developers enjoyed yesteryear.
The fact is that much of that framework craze is pure hype. And the java web development crowd should testify to that: how many frameworks du-jour have seen the ligth to be abandoned a couple of months / years later? JSF, javabeans, portlets, you name it. All the while quick-n-dirty php conquered the web.
IMHO the main error is trying to abstract too much out of the HTTP(+HTML+CSS) model: OOP guys love to pretend they're coding inside a GUI app, and let the magical framework do the translation for them. Then one day they wake up and the realize that either: for drawing one html page that framework issues 1000 ajax calls - bad for scalability, customer gets real pissed, project canned, or: that shiny new css property that would allow page design to be 100% perfect is not part of the framework, sorry dude, or: no way to update only part of your web page via ajax without completely breaking the framework - it's his job to compose the page, after all, or: the html the framework spits is not firefox compliant, etc...
Keep It Simple folks!
php is dead easy for the web because it follows spot-on the web model. One html page = one php script. GET, POST, COOKIES always accessible within your app. Stick as little data into the session as you can. Let the VM clean up at the end of every single request all junk the programmer allocated. etc...
AFAIK, delphi for delphi had this kind of design-with-a-gui-we-translate-to-html power years ago, and I've never seen a single web app deployed like that..Net takes the same path, and has not been booming, except inside all-ms shops.
Plus, people that do not know coding use flash already. Little value added in switching.
As far as I'm concerned, PHP would also be a good contender for "ideal scripting language in the browser".
Advantages over js: - using the same lang both server and client side is easier on the coder (applies to python too, and might apply to js, if it ever becomes widespread server side) - huge set of functions in the interpreter core: I really miss a lot of them when coding in javascript - the object model, albeit more restrictive than the prototype based model, makes a lot more sense to people used to C++, java, php et al
Dude, I thought this was a guide on optimizing ORACLE performance! Its sums up neatly all the advice I keep giving to all the noob contractors that develop on sqlserver and then kinda start having nightmares when they are told we will install on oracle in production (and their app gets suddenly 5x slower...)
Slashdot Mirror
Yeah, I was prototyping this in Milano airports as well, around 2006.
Using RFIDs instead of barcodes on luggage tags makes a lot of sense, as you can store within the chip itself the complete story of its journey, instead of having to store it in airport databases, and later retrieve it wherever the bag is. The dream scenario was to store your hotel address on the bag and have it delivered for you all the way there.
Also, RFID readers can scan a few hundred chips or more at a time, so it is easy to install them over doors hand have them check automatically everything that goes through.
The project was a great success from a technology standpoint, but it did not take off because:
- it only works good if many airports adopt RFID, not just one
- RFID chips at the time cost 1 eur apiece, much more than printed barcodes (and there are *a lot* of bags travelling on any single day)
Nice to see that things are catching up 10 years later
I see lots of comments trying to shift the blame from the administration that originally dumped the chemicals in what was at the time countryside.
Funnily not a lot of them highlighting the fact that dumping toxic waste, wherever it might be, is just plain BAD. And this is a textbook case: you might never know what will happen there in 50 years, let alone 500.
Food for thought for nuclear waste dumps...
Was it the american indians that encouraged to think about the consequences of one's actions for seven generations?
Totally agree - and yet naming the eZ Components that way lost 50% marketshare to the zend framework on day one. Talk about fads not existing in the it world...
ESA just launched a new website for hosting videos - check it out
http://multimedia.esa.int/content/search?SearchText=iss&SearchButton=Go
http://multimedia.esa.int/content/search?SearchText=mars&SearchButton=Go
100% agree. Install filemon,diskmon,regmon,portmon and off you go. Install Process explorer as replacement for the rask manager. Read MR's blog to learn how the big boys go at it. Look for hints at the event viewer: any suspicious/recurring alert in there?
Explained better than I could ever do: http://www.well.com/~doctorow/metacrap.htm
Hmmmm...
Little hands-on experience with eee 901 xandros and wife:
- eee is very well set, up, hard to tell it is linux, and the gui is 100% locked up: good for average joe, but wife wanted to have thunderbird on 1st menu page, not 3rd - and off we go with wikis and the usual linux do-it-yoursel approach
- wife was very happy with games: more than windows, and coming witha tutorial, too!
- firefox, skype: no problem in there after I set up the accounts (easy: she used the same sw on win)
- ooo: complete failure. Being a math teacher, she relies heavily on inserting formulas in docs. The interface in writer might be a bazillion times better than in word, she still did not want to learn it. And all the options to disable automatic reformatting of text are not in the same menus, so disabling them also proved to be quite hard.
In short: the devil is in the details - every average user has a few specialized needs and expectations. Getting everything right - and making it as similar to windows as possible at the same time is an impossible task
in short, are you advocating usage of virtuozzo?
That was about the first question that came to my mind: are the Python/Parrot/PHP/Ruby guys aware of these developments in js technology? Are they already porting the techniques? Or is it impossible to take advantage of them because of the basic assumptions in the underlying VMs?
After all, despite all these languages interpreters being mature enough (erm, maybe the P ones...), javascript engines are seeing a tremendous advance in speed after many years of relative stagnation...
Was it a tool by CAST software, by any chance?
The grand-unified-find-interrelations-between-your-sql-and-your-code was the most over hyped useless garbage I have met in a while.
It eg. found in delphi code that a button named the same as a db column was somehow tied to the db.
Or in php code, it was not even able to tell what KIND of databases the code would connect to, since db connections depended on ini files, and it could not grok includes based on variables.
Talk about some false positives...
In Italy, newspapers do the "omfg! we are running out of engineers! 50.000 indians or chinese will need to move to italy to compensate shortage!" article at least every couple of years.
The fact is, there are a lot of companies that want 50.000 engineers paid 800 EUR a month. And a lot of engineers that have a very hard time finding a job that rewards them 2.000 euros.
Nuff said (but the uncoolness of studying math or physics is biting hard universities, and in the engineering degrees all the pupils only go for the management-oriented curricula. This is going to prove a problem in a couple of years if not further away)
way to go, bofh spirit. And what if the client is a better dba than you and has a very precise understanding of roles, grants and privilege separations in Oracle?
Your description really looks like the plot for the Ranxerox comic - where a few years in the future this huge, exceedingly brutal, dope addict robot is called in to hollywood to play Fred Astaire in what is planned to be the biggest remake of all time...
BTW hey, nobody ever made a Ranxerox comic! Studios, do you hear me? Here's the next movie idea served to you on a plate!
Sorry to bust your bubble, but shared bikes are all the rage in France right now.
Lyon has them, Paris too, and all other biggish cities are scrambling to get theirs.
All it takes is an almost-civilized nation and some careful execution.
The shared bikes-done-right involves:
- you only ride them for free for 1/2 an hour, afterwards it's paying
- this means you have alot of incentive to never go with the bike too far away from a parking lot
- about a hundred thousand parking lots covering the complete city
- at every parking lot a pc tower has a map with the nearest other parking lots
- bikes can only be rented after you sign up for your rental plan, at no cost but involving a caution and a bank account
- bike parking lots are web connected, you can find their placement on the web and even reserve one bike for when you need it
In paris it has been going well, and the city life is though, by all accounts. Bikes stolen/broken/vandalized have been in the range expected by the programme (ie there is some cash allotted from day one for repairs / refurbishing)
Everybody knows how the story goes: after mighty mouse experiments, both Russia ans the states turn to secret experiments to build human superheroes for the army, and it ends up in WW3, the world being destroyed and a new kind of humanity in search of a new planet: The One
Rick Veitch saw this coming, twenty years ago!
Yes, there indeed are laws regulating the 'printed' press, and it is the main justification for extending the law: why should slander/libel be condoned on a highly visible blog more than it is on a newspaper selling a few hundred copies?
Also note that Italy has a dumb anti-terrorism law that basically forbids everybody but the telcos to provide free wifi access (people who do MUST take a copy of the id card of every wifi user that connects to their hotspot, and keep logs of all web access), and it is really hurting the mobile workforce.
US is not the only government with police-state temptations that needs to be kept in check...
Not bank related, and not in the US, but this happened to me: ...and they charged me 150eur for the favor (the price of a new connection)! They stated it was my duty to recover money from 'the other company' ...but all other operators told me that they could not provide me a contract using the same telephone number, because the original company would never surrender it, since I had some debt pending (the fine I refused to pay). Note that they had already surrendered the number in the first place, breaching in fact their contract with me, but they would refuse to do it when asked.
- telco cut my phone cable, without any previous warning, because of what amounted to an accounting error on their part
- I had to jump through loops for a week before telco admitted their fault. Initially they blamed it on competing companies requesting to switch my number over, and competing telcos denied that.
Note that all of the telcos only have customer desks that respond via call-center: not very easy to call when you've got no phone (tip: calling call centers via mobile costs lottsa money)
- after admitting their error (even though the disconnect request in fact did came from a competitor), they gave me the connection back - within a month...
-
- I refused to accept that logic (THEIR FAULT, I HAVE TO SUFFER), and decided to switch operator...
-
This long post just to clarify one thing: you do not always have the choice to change banks/telephone/services providers. And the logic 'customer always at fault' does indeed apply 95% of the times
Not all airport display systems are created equal, but to use a full fledged PC with a MS os onboard is not the only solution.
For biggish deployments, every monitor usually comes with its own single-motherboard-industrial-pc in the casing, possibly running some sort of embedded linux distro and either running a custom app that receives raw data from the central server, formats it and displays it locally or just running an xserver to drive the display with the application running on said server. Similar setups (citrix/terminal services on wince) are possible in windoze land.
In such cases, there is no restriction on cabling, since transport is standard ethernet, and very little cost saving...
I guess different people have different learning approaches, so what is perceived as "easy" or natural by someone is deemed obscure and arcane by someone else. In general, I prefer to read the 'technical specification' manual rather than the 'filled with examples' tutorial, but I understand that in order to get any benefit out of it you need at least a basic understanding of the topic at hand.
/. readers have duly pointed out all the many advantages that apache has (such as using a simple search command to find out if a parameter exists/has been set: it can be a real time saver compared to the infinite series of tabs/dialogs that have to be navigated with IIS when you are not really sure where things are set)
More to the point:
- I find Apache docs very good, both when compared to many other OSS projects and when evaluated on their own. The basic params are described one by one, and you can usually find the correct combination with a little trial and error, but for the most complex tasks there are dedicated sections which are very helpful (eg tuning)
- 'using a text editor' to edit the config file has a very nice side consequence: it makes backing up the config and reimporting it later / duplicating it extremely easy. But I am sure a lot of other
- by your own admission, you have never configured a moderately complex website. I would consider the fact that you do not know what http compression is a sore spot in your CV, but that is just my own personal opinion. IMHO the good coders are the ones that have a passion for learning (at least at the basic level) all the aspects involved in their work, be it system administration, network protocols or memory allocation / cpu architecture.
- to be honest, IIS is indeed more friendly 'out of the box' to the complete newbie. It complements that with the sheer inability to implement all of the complex setups that apache can do, and with an exponential growth of the time needed to configure it vs. configuration complexity (the example about setting folder perms being a prime example: it's fine and dandy as long as you develop your small app over which you have complete control, but just imagine having to manage tens of different web applications developed by different people, that do not write down clear specs for configuration and have very sloppy security practices, and you see the mess it becomes. Last week I had to deploy an app. where the file containing the db credentials - in clear was set to be world-readable... I was NOT pleased).
Otoh Apache has a higher barrier to entry, but it gets easier to use the more you know it. This is a well known paradigm for both OSS (makes easy the hard things, but not easier the easy ones) and microsoft sw, which makes extremely easy the easy things and impossible the complex ones.
Bye
Gaetano
Ok, so did you also try to:
- read the apache manual that comes with the distro instead of googling around as 1st stop help shop
- add to logfiles a custom http request header (requires writing a custom dll in IIS)
- have different logs generated by different virtual hosts / folders / requests
- edit by using a text editor the iis config / back it up or restore it / write a plugin to manipulate it (impossible up to iis 6)
- use ldap as authentication backend for IIS
- find out the access perms set on a huge tree of IIS folders without clicking them open one by one
- set up decent http compression in IIS without resorting to external tools such as Port80
- set up IIS as reverse proxy or load balancing proxy
sorry, but as far as ease of use, configurability and sheer number of features there really is no match. Apache can be tweaked to serve every possible imaginable unconceivable convoluted application you design. IIS cannot.
Absolutely agree with parent.
.Net takes the same path, and has not been booming, except inside all-ms shops.
Lots of people nowadays will be happy to tell you how there are 100 (free) MVC php frameworks and templating systems out there that allow "proper" enterprise web app coding instead of the ugly db-logic-presentation mixup that php developers enjoyed yesteryear.
The fact is that much of that framework craze is pure hype. And the java web development crowd should testify to that: how many frameworks du-jour have seen the ligth to be abandoned a couple of months / years later? JSF, javabeans, portlets, you name it.
All the while quick-n-dirty php conquered the web.
IMHO the main error is trying to abstract too much out of the HTTP(+HTML+CSS) model: OOP guys love to pretend they're coding inside a GUI app, and let the magical framework do the translation for them. Then one day they wake up and the realize that either: for drawing one html page that framework issues 1000 ajax calls - bad for scalability, customer gets real pissed, project canned, or: that shiny new css property that would allow page design to be 100% perfect is not part of the framework, sorry dude, or: no way to update only part of your web page via ajax without completely breaking the framework - it's his job to compose the page, after all, or: the html the framework spits is not firefox compliant, etc...
Keep It Simple folks!
php is dead easy for the web because it follows spot-on the web model. One html page = one php script. GET, POST, COOKIES always accessible within your app. Stick as little data into the session as you can. Let the VM clean up at the end of every single request all junk the programmer allocated. etc...
AFAIK, delphi for delphi had this kind of design-with-a-gui-we-translate-to-html power years ago, and I've never seen a single web app deployed like that.
Plus, people that do not know coding use flash already. Little value added in switching.
Best essay on the topic I have come across: http://www.well.com/~doctorow/metacrap.htm
As far as I'm concerned, PHP would also be a good contender for "ideal scripting language in the browser".
Advantages over js:
- using the same lang both server and client side is easier on the coder (applies to python too, and might apply to js, if it ever becomes widespread server side)
- huge set of functions in the interpreter core: I really miss a lot of them when coding in javascript
- the object model, albeit more restrictive than the prototype based model, makes a lot more sense to people used to C++, java, php et al
Dude, I thought this was a guide on optimizing ORACLE performance!
Its sums up neatly all the advice I keep giving to all the noob contractors that develop on sqlserver and then kinda start having nightmares when they are told we will install on oracle in production (and their app gets suddenly 5x slower...)