I definately agree with you, and I'll be the first to throw a fit when that happens (lately though I'm seeing more and more of the opposite. It is still rare, but FF/Chrome/Safari-only web sites are starting to pop up... The irony is that in some cases, those sites make use of things where those browsers are not standard compliant: since they tend to use each other to validate themselves, instead of validating against the actual standard, it happens more than you'd think...fortunately its still rare).
Anyhow, was just pointing out that Silverlight is most definately not IE-only, or even Windows-only for that matter. It -is- officially supported even on Macs, and while it is only indirectly supported in Linux via Moonlight, microsoft themselves will redirect you to Moonlight's (open source! not even made by Microsoft!) web site if you try to access silverlight-content in Linux.
Anti-spyware/malware maybe. Firewall rules however, useless. What do you do if the software simply does an HTTP Post to a web service with default proxy settings, or if the sysadmin is clever and uses a setup where the default proxy settings are not being used, its not too hard to sniff/autodetect them. Not much to do when the malwares use the same outbound as another important piece of software...
He didn't say security software. He talked about security measures in general. That means making users have strong passwords and change them regularly, not running every application they see, etc.
You can spout how Linux is more secure by design all day, but usually the current user context is MORE than enough to do damage, no matter how restricted you make it (if a user can read their own email and type in their own browser and read their own instant messages, so can a software, and its all whats needed). There's nothing you can do when a hacker can just CALL the damn user, ask them for their password, and they'll hand it over. So you need to add VPNs, proxies, restrict internet access, force password changes, etc, no matter which OS. But that is met with extreme resistance (hell, even IT people, even I, will resist it...). The software is rarely the problem.
Im not too sure why you bring up Silverlight and in the next sentence talk about needing to switch from Firefox to IE though. Not like silverlight is IE (or even windows) centric.
Many more data types, including a datetime type that doesn't suck. A new datatype that acts like it stores files in the database but keeps them in an external folder (that one is probably one of the biggest. This is really a big deal. Much more powerful than rolling up your own solution). Major enhancements of the devtools (native SQL intellisense). ETL and business intelligence stuff got big improvements (ability to use any.NET language in SSIS and better use of multi-core CPU for ETL, enhancements to SSAS, etc). New file formats in reporting services. Much better query planner. All around performance improvements.
The list goes on. I'd say 2005 -> 2008 is a bigger jump than 2000 to 2005. At first glance it isn't, because they didn't rewrite as many parts from scratch (like SSIS and Management Studio), but the new datatypes, enhancements to T-SQL, the souped up business intelligence suite, external file management, and performance improvements are a much bigger leap.
Of course. But then again, 2008 is leaps and bounds beyond 2005 too.
2000 was good, but paled compared to some of the more expensive competition, and lacked some "enterprise" features. 2005 was a competitive offering. 2008 is in many ways ahead of the competition. And the dev tools rock:)
Educational pricing aside, thats always how commercial software works when it comes from the big companies (Microsoft, Oracle, etc).
Take MSDN Subscriptions for example.
1 year subscription, 2000$~ (Im making numbers up, but the proportion is accurate) 2 year subscription, 2250$~ 3 year subscription, 2350$
See where this is going?
Where i'm working for is a corporation that is not educational or charity in nature (nowhere close...making billions in profit), and Microsoft sells us Windows and Office for something silly like a few douzan dollars per seat for all of it together, including sharepoint enterprise CAL and whatsnot, and we only pay at the end of a given period of time (so if someone gets hired and then quit before the period is up, we dont pay for them).
Except that Ultimate is not targeted to the average user, and the vast majority of people who would be interested in it will have access to some form of discount or another.
And if you buy a computer with one of the other editions preinstalled, it will be a fraction of the price, and netbooks will qualify for the super cheap edition.
The rest will qualify for upgrades.
People who build their own will be able to use OEM pricing.
This basically only affects people who build PCs from spare parts at Best Buy. All 3 of them.
I don't know about MySQL, but prepared statements in most major RDBMs will allow the database to cache the query plan as well as being more easily optimized. So they actually are much -faster- if you need to execute the query over and over (especially if you can reuse the same query object). Many database APIs will also let you use statement objects that have the same capabilities as prepared statements in term of query plan caching and safety, but do not do the first roundtrip to optimize the query on the server, so if you are just going to execue the query once, you can still use a parameterized query without the double roundtrip.
ORM (Object relational mappers) will use these methods internally and are fairly pervasive in high performance application.
And its not wasteful even if you control the sql, again, because of the query plan caching abilities that it provides.
The only time I ever need to reboot a windows desktop is for kernel patches. While it IS significantly easier to apply kernel patches in Linux without rebooting, its still a pain and I don't know many Linux users who'll do that without rebooting, either.
Again, this isn't 1998 anymore. Windows, OSX, Linux...neither crash, neither need reboots, like, ever (aside for the above exception, which I'll admit is a weakness of Windows, since you'll patch on average exactly once a month). In either case, only faulty drivers (which isn't uncommon on Linux either...) and bad hardware will bring either down.
Yeah, and I don't think you know the meaning of "more", eh? And if you have a fucking server doing exactly the same thing over and over and over, if it can run for a week without any hiccup, it can run for a year. Especially when talking about memory: if memory usage didnt spike in a month, it won't in 12-24 months either.
Even in Windows-land, the bottleneck now is hardware reliability. Which was my point.
I could point at the many memory "leaks" found in Windows, which MS has spent a lot of time trying to fix. Maybe that's all fixed in Win7 - but I have my doubts. To my knowledge, no other OS suffers from that problem, including Linux. Some few applications that run on Linux suffer from "leaks", but Linux handles that all differently, so we never end up with a gigabyte of memory dedicated to some process that is totally unproductive and unresponsive
Look at the AC post that explained the issue with what you said better than I will... on my side, I'll just point out that this isn't pre-2000 anymore. You can have Windows running (and actively doing stuff) for hundreds of hours (or more) without ever seeing anything quirky anymore. And aside for the odd case of bad hardware, or in the few occasions I risked beta drivers, I havent seen a blue screen in a little under a decade.
But, Linux manages memory completely differently than Windows does. Linux will generally fill memory up, as much as possible, instead of swapping out to disk.
Thats what Windows does too, and as someone pointed out already, goes a step further and precache stuff in advance, especially on Vista, and is the behavior responsible for people flipping out about how bloated it is (then they just hide it on Win7 and everyone is in awe at the amazing efficiency machine that is Win7...lol).
Apache and IIS have had pretty much no remotely exploitable holes in years (in the core. Of course some extensions and modules had issues), so whenever either have one found, its news.
The only issue (and don't get me wrong, I love IIS and its my primary web server, for both personal and professional use) is that Microsoft's web server only exposes a fraction of its options through the GUI... when things get heated, and you have to actually drop to the config file, IIS is a serious nightmare.
At like 10 seats SA makes sense. Less than that if you want actual support. And thats if you only get Windows. If you plan on getting other MS products, you can get benefits at as little as 2 seats.
Hell, im on SA as a single developer, because the support contract and the savings made it cheaper than the alternatives...
Its a little more than that, since the app can work offline, and detect in which context its working (so you can make the app work differently when its in a browser and when its stand alone: the described scenario example is to have you save data locally in a sandbox when offline, and synchronize with the server when online...there's tools provided to do it).
You're right though: its not exactly complex. Its just a good idea. And in 2009, good ideas are harder to come by than technology:) Thats true for most fields, too.
No, this is not the same thing. Webstart is like.NET's ClickOnce (which, admitedly, came in later).
What Silverlight 3 does is let you make a browser RIA that the user can "detach", and move and use locally, with or without an internet connection. So basically imagine if you had an applet that could become a WebStart app, with virtually zero effort on the developer side, and completly streamlined on the user site.
Flash and Silverlight 1, yes, you're right. Flex and Silverlight, not so much: they're just not browser APIs, but actual app frameworks, so they go a lot further than just providing end user features: they help the programmer, too (like Silverlight can consume an ADO.NET data service with LINQ, which is a lot better than using the built stuff, even in HTML 5. Flex has BlazeDS, and so on. Just an example)
JNLP is closer to.NET ClickOnce (actually its the same thing, just for.NET. And yeah, ClickOnce came in later). The other.NET technologies that need to run in a similar sandbox just tap into the same APIs. So XBAP apps and Silverlight, for example.
Slashdot Mirror
I definately agree with you, and I'll be the first to throw a fit when that happens (lately though I'm seeing more and more of the opposite. It is still rare, but FF/Chrome/Safari-only web sites are starting to pop up... The irony is that in some cases, those sites make use of things where those browsers are not standard compliant: since they tend to use each other to validate themselves, instead of validating against the actual standard, it happens more than you'd think...fortunately its still rare).
Anyhow, was just pointing out that Silverlight is most definately not IE-only, or even Windows-only for that matter. It -is- officially supported even on Macs, and while it is only indirectly supported in Linux via Moonlight, microsoft themselves will redirect you to Moonlight's (open source! not even made by Microsoft!) web site if you try to access silverlight-content in Linux.
Anti-spyware/malware maybe. Firewall rules however, useless. What do you do if the software simply does an HTTP Post to a web service with default proxy settings, or if the sysadmin is clever and uses a setup where the default proxy settings are not being used, its not too hard to sniff/autodetect them. Not much to do when the malwares use the same outbound as another important piece of software...
He didn't say security software. He talked about security measures in general. That means making users have strong passwords and change them regularly, not running every application they see, etc.
You can spout how Linux is more secure by design all day, but usually the current user context is MORE than enough to do damage, no matter how restricted you make it (if a user can read their own email and type in their own browser and read their own instant messages, so can a software, and its all whats needed). There's nothing you can do when a hacker can just CALL the damn user, ask them for their password, and they'll hand it over. So you need to add VPNs, proxies, restrict internet access, force password changes, etc, no matter which OS. But that is met with extreme resistance (hell, even IT people, even I, will resist it...). The software is rarely the problem.
Im not too sure why you bring up Silverlight and in the next sentence talk about needing to switch from Firefox to IE though. Not like silverlight is IE (or even windows) centric.
Many more data types, including a datetime type that doesn't suck. A new datatype that acts like it stores files in the database but keeps them in an external folder (that one is probably one of the biggest. This is really a big deal. Much more powerful than rolling up your own solution). Major enhancements of the devtools (native SQL intellisense). ETL and business intelligence stuff got big improvements (ability to use any .NET language in SSIS and better use of multi-core CPU for ETL, enhancements to SSAS, etc). New file formats in reporting services. Much better query planner. All around performance improvements.
The list goes on. I'd say 2005 -> 2008 is a bigger jump than 2000 to 2005. At first glance it isn't, because they didn't rewrite as many parts from scratch (like SSIS and Management Studio), but the new datatypes, enhancements to T-SQL, the souped up business intelligence suite, external file management, and performance improvements are a much bigger leap.
Of course. But then again, 2008 is leaps and bounds beyond 2005 too.
2000 was good, but paled compared to some of the more expensive competition, and lacked some "enterprise" features. 2005 was a competitive offering. 2008 is in many ways ahead of the competition. And the dev tools rock :)
Educational pricing aside, thats always how commercial software works when it comes from the big companies (Microsoft, Oracle, etc).
Take MSDN Subscriptions for example.
1 year subscription, 2000$~ (Im making numbers up, but the proportion is accurate)
2 year subscription, 2250$~
3 year subscription, 2350$
See where this is going?
Where i'm working for is a corporation that is not educational or charity in nature (nowhere close...making billions in profit), and Microsoft sells us Windows and Office for something silly like a few douzan dollars per seat for all of it together, including sharepoint enterprise CAL and whatsnot, and we only pay at the end of a given period of time (so if someone gets hired and then quit before the period is up, we dont pay for them).
Thats why early upgrades will be half price. Google it. (not kidding)
Except that Ultimate is not targeted to the average user, and the vast majority of people who would be interested in it will have access to some form of discount or another.
And if you buy a computer with one of the other editions preinstalled, it will be a fraction of the price, and netbooks will qualify for the super cheap edition.
The rest will qualify for upgrades.
People who build their own will be able to use OEM pricing.
This basically only affects people who build PCs from spare parts at Best Buy. All 3 of them.
I don't know about MySQL, but prepared statements in most major RDBMs will allow the database to cache the query plan as well as being more easily optimized. So they actually are much -faster- if you need to execute the query over and over (especially if you can reuse the same query object). Many database APIs will also let you use statement objects that have the same capabilities as prepared statements in term of query plan caching and safety, but do not do the first roundtrip to optimize the query on the server, so if you are just going to execue the query once, you can still use a parameterized query without the double roundtrip.
ORM (Object relational mappers) will use these methods internally and are fairly pervasive in high performance application.
And its not wasteful even if you control the sql, again, because of the query plan caching abilities that it provides.
The only time I ever need to reboot a windows desktop is for kernel patches. While it IS significantly easier to apply kernel patches in Linux without rebooting, its still a pain and I don't know many Linux users who'll do that without rebooting, either.
Again, this isn't 1998 anymore. Windows, OSX, Linux...neither crash, neither need reboots, like, ever (aside for the above exception, which I'll admit is a weakness of Windows, since you'll patch on average exactly once a month). In either case, only faulty drivers (which isn't uncommon on Linux either...) and bad hardware will bring either down.
Yeah, and I don't think you know the meaning of "more", eh? And if you have a fucking server doing exactly the same thing over and over and over, if it can run for a week without any hiccup, it can run for a year. Especially when talking about memory: if memory usage didnt spike in a month, it won't in 12-24 months either.
Even in Windows-land, the bottleneck now is hardware reliability. Which was my point.
Look at the AC post that explained the issue with what you said better than I will... on my side, I'll just point out that this isn't pre-2000 anymore. You can have Windows running (and actively doing stuff) for hundreds of hours (or more) without ever seeing anything quirky anymore. And aside for the odd case of bad hardware, or in the few occasions I risked beta drivers, I havent seen a blue screen in a little under a decade.
Thats what Windows does too, and as someone pointed out already, goes a step further and precache stuff in advance, especially on Vista, and is the behavior responsible for people flipping out about how bloated it is (then they just hide it on Win7 and everyone is in awe at the amazing efficiency machine that is Win7...lol).
Ok, 50 megs then. But most of that memory is probably shared,so I'd guess its actually quite a bit less.
27.8Mb here...
Apache and IIS have had pretty much no remotely exploitable holes in years (in the core. Of course some extensions and modules had issues), so whenever either have one found, its news.
The only issue (and don't get me wrong, I love IIS and its my primary web server, for both personal and professional use) is that Microsoft's web server only exposes a fraction of its options through the GUI... when things get heated, and you have to actually drop to the config file, IIS is a serious nightmare.
Which is less than half retail as soon as you buy like 3 licenses under SA.
At like 10 seats SA makes sense. Less than that if you want actual support. And thats if you only get Windows. If you plan on getting other MS products, you can get benefits at as little as 2 seats.
Hell, im on SA as a single developer, because the support contract and the savings made it cheaper than the alternatives...
Its a little more than that, since the app can work offline, and detect in which context its working (so you can make the app work differently when its in a browser and when its stand alone: the described scenario example is to have you save data locally in a sandbox when offline, and synchronize with the server when online...there's tools provided to do it).
You're right though: its not exactly complex. Its just a good idea. And in 2009, good ideas are harder to come by than technology :) Thats true for most fields, too.
No, this is not the same thing. Webstart is like .NET's ClickOnce (which, admitedly, came in later).
What Silverlight 3 does is let you make a browser RIA that the user can "detach", and move and use locally, with or without an internet connection. So basically imagine if you had an applet that could become a WebStart app, with virtually zero effort on the developer side, and completly streamlined on the user site.
Flash and Silverlight 1, yes, you're right. Flex and Silverlight, not so much: they're just not browser APIs, but actual app frameworks, so they go a lot further than just providing end user features: they help the programmer, too (like Silverlight can consume an ADO.NET data service with LINQ, which is a lot better than using the built stuff, even in HTML 5. Flex has BlazeDS, and so on. Just an example)
JNLP is closer to .NET ClickOnce (actually its the same thing, just for .NET. And yeah, ClickOnce came in later). The other .NET technologies that need to run in a similar sandbox just tap into the same APIs. So XBAP apps and Silverlight, for example.
You had a point until there.