This may suprise you, but my experience with real projects has been that the bug count *does* reduce to some sort of equilibrium, then it seems to stabilize. Bugs arrive at the same rate they are fixed.
What is happening is that the character of the bugs is changing - people are no longer finding grade A 'failure' type bugs, but instead are beginning to focus on lower priority and cosmetics bugs. Once a project reaches this stable point, it seems to stay there for a long time (or at least long enough that we decided to release new features and mess it all up again).
Doesn't have to be hosted on Google. If they produce an ajax office suite - it could be open source and run on the corporate server. Suddenly office PCs need only be a browser!
There are very few xerox printers that actually work (Xerox is a mess internally and almost every printer has incompatible commands with every other printer).
Lexmark printers are well covered - most printers share a common set of commands. An N16 will work for an N32, an optra R will work for optra {R,S}* and so on.
HP printers are reasonably well represented, and share common commands.
I've really tried to love linux in the corp world, but until I can actually print a duplex rotated document to my high-end MFD, its a no go.
graphical app (1/2 of a CHECK). Gimp UI devs need a sharp rap across the knuckles. Otherwise, it would be a CHECK,
dont require users to dive into config files and command lines (e.g. wpa anyone??). Every app should have a GUI admin. All config should be done through a control panel (collection of admin tools).
having 100 possible text editors is not something to be proud of. Pick one. Vi (on unixes in the 80s) and notepad today are success stories - consisent, everywhere, something you can count on. (Devs can do what they like, this is users I am talking about).
Fix the 'start' menu. Its like every distro tool the MS start menu and messed it up. Look at Ubuntu if you want to see menus organised right.
I have used redhat, suse, fedora and ubuntu. Ubuntu kicks butt as far as usability. Simple install plus one of everything right where you expect. 5.10 is excellent with OO2.
As the parent says, MCE is a performance hog, which means you need a grunty CPU, which means heat and lots of it. Heat requires cooling, and cooling is the prime cause of noise.
Trust me - you dont want a noisy PC in your living room.
It *is* possible to reduce cooling noise significantly in a box running MCE, but the price is driven up, and you often end up having to customise the box. Further, there are likely to be water pumps or fans still in there.
Myth is a *pig* to install (even with Knoppmyth, which I use), but you can build a completely silent system.
Dont underestimate how much the noise will bother you.
using MythTV. Strips them out automatically. Sadly misses the odd one, but I have 'skip30' and 'back5' buttons on my remote to solve that - 7 or 8 quick clicks past the ads, then back to the start of the prog.
I haven't seen an ad in many months. TV has improved out of sight for me.
This is a very good question - and as always it comes down to the best tool for the job.
IAADBD (I am a DB designer) who has designed and implemented DBs for large apps (millions of rows, hundreds of entities). If your data is important, you need to look at this from the data's point of view.
Ensure the data stays clean
Ensure the data is well designed for growth (normalization)
Execute business rules inside the DB
I achieve these by permitting access only via stored procedures (with rare, simple exceptions), and having complete coverage with DRI (declaritive referential integrity), triggers, defaults, rules, timestamp checking and so on. I control what access people/processes have to the data, and am able to encapsulate people away from the underlying structures.
I can therefore be sure the data stays clean and well formed. I can also be sure that people are not pulling large gobs of data out of the database and across the network to implement business logic that should be done inside the DB.
If your data can be adequately protected using DRI alone, and you are not so concerned with encapsulation and performance then these tools are brilliant.
Exactly! This question points to a huge misunderstanding with MySQL. So called 'industrial strength' RDBMSs are very vary careful to ensure your data is 100% correct. These RDBMSs manage very large databases with GB of data. If (e.g.) referential integrity is lost, you are sunk. If data corruption sneaks in, its pretty hard to fix it or find it - there's way too much data and complexity.
So these RDBMSs provide tools - triggers, stored procedures, a multitude of locking mechanisms, DRI (Declaritive Referential Integrity) and so on. With a well built database:
You *can't* insert duplicate data
You *never* modify data behind a users back (e.g. remove trailing spaces on varchar)
You *never* break a foreign key, leaving dangling data
You can perform all maintenance while the DB is live
and so on...
The SQL standard is very clear and explicit about data handling - in general no surprises and never let in bad data. And all 'industrial strength' RDBMSs do a good job of following the standard... sadly not perfect but very good. MySQL is not one of them - see here for some of the ugliness inside MySQL exposed. For example: MySQL's use of NULL is just bizarre, as is its ability to accept invalid data and change it silently.
'Industrial Strength' RDBMSs are about predicability and reliability. Speed is important, correctness is essential. MySQL aint there yet. Oracle, MSSQL, Postgres, Sybase, to name a few of my favourites, are.
IMHO If you wrote the code on their time, then they should own all rights to the code. If they elect to give you some rights, great. If they elect to to GPL the code, awesome.
If you had some libraries that you wrote in your time and brought it, you will want to keep them.
Actually (small correction) - I have an 80W power supply.
I have a full sized hard drive, DVD burner and an encoder/decoder card (Hauppauge PVR-350), external HDD via firewire.
With the 60W and everything loaded, I had intermittent failures on start. When I put the 80 on, I found that it drew about 72W very briefly after start. Its now reliable.
Eventually he will step over some line and annoy SteveB. At that point, MS will sue and subpoena logs from blogspot. IP address and personal details will quickly be revealed.
Kinda makes you wish for true anonymity in blogging that a tool like freenet could provide (if it weren't such a dog slow pile of dung).
The fact that this is such a fast moving standard is very interesting. Most standards take years of negotiation to move forward - with MS at the helm this one is fair trucking along.
In effect, this makes the standard a moot point - it is unlikely that anyone will have the resources or willpower to create a competitive implementation to C# given the goal posts move so rapidly.
MS have brilliantly created a standard, and probably what will be the only real-world usable implementation of that language. They get the best of both worlds - standards and proprietary. Cool.
Even if the quality is not the best, those of us who pay up to several dollars a minute for cellphone use will find it pretty useful.
I am roaming with GSM - and am thinking of direct crediting my pay straight to vodaphone to save on bank charges. Its just ridiculous - every call (even local to where I am) is charged as a toll call.
I can only imagine mikejz84 is an American who does not travel overseas very much. Local calls in the US are very cheap, wireless plans often build in hundreds (or thousands) of minutes per month. Once you leave Oz, its not such a pretty story.
Also, in the last two years the linux kernel has been spiraling out of control.
In the past we had the 2. stable branch. Stable, bug fixes, few new features. Then we had the wild west of the 2, branch with a pell-mell of features, bugs and all that good stuff that goes into a fast moving product.
Now we just have the 2.6 tree. Its got all the good stuff but its also the dev branch. Its moving like crazy, stuff is breaking (USB) and being fixed at a great rate.
If you are building a major server, you want a stable platform. You dont want kernel updates every few weeks 2.6.11.1,.7, 2.6.12.xxx, 2.6.13 just out and so on.
I think the move away from odd + even branches has been a net loss to the reputation of the linux kernel.
Another problem is that satellites are affected by numerous gravitational sources (e.g. moon, sun) and slowely wonder off course. For example, equatorial satellites require small adjustments every year. While their mass is very low - there didn't look to be a lot of room inside that box for some thrusters.
Its not all rosy. The kernel version (2.6.11.7) used in Knoppmyth has an issue with USB hard drives - hence my extra HDDs plugged into the myth box were attached using firewire (not that its a problem).
My point is that the 2.6 kernel is not without its flaws - it depends which particular kernel version you get and which warts it has.
My advice is if it ain't broke...
Do you really *need* the hassle of the upgrade?
What is happening is that the character of the bugs is changing - people are no longer finding grade A 'failure' type bugs, but instead are beginning to focus on lower priority and cosmetics bugs. Once a project reaches this stable point, it seems to stay there for a long time (or at least long enough that we decided to release new features and mess it all up again).
Nope. The windows version of OO.o is 76,756KB.
Wait... haven't we been here before.
There are very few xerox printers that actually work (Xerox is a mess internally and almost every printer has incompatible commands with every other printer).
Lexmark printers are well covered - most printers share a common set of commands. An N16 will work for an N32, an optra R will work for optra {R,S}* and so on.
HP printers are reasonably well represented, and share common commands.
I've really tried to love linux in the corp world, but until I can actually print a duplex rotated document to my high-end MFD, its a no go.
Further, the whole thing started at the time he got sued by the govt. It was originally a PR exercise.
Still, it is doing some good work.
This is the ultimate chick flick.
This is the most estrogen filled topic yet.
I have used redhat, suse, fedora and ubuntu. Ubuntu kicks butt as far as usability. Simple install plus one of everything right where you expect. 5.10 is excellent with OO2.
As the parent says, MCE is a performance hog, which means you need a grunty CPU, which means heat and lots of it. Heat requires cooling, and cooling is the prime cause of noise.
Trust me - you dont want a noisy PC in your living room.
Myth runs very well on low spec fanless EPIA mobos.
It *is* possible to reduce cooling noise significantly in a box running MCE, but the price is driven up, and you often end up having to customise the box. Further, there are likely to be water pumps or fans still in there.
Myth is a *pig* to install (even with Knoppmyth, which I use), but you can build a completely silent system.
Dont underestimate how much the noise will bother you.
using MythTV. Strips them out automatically. Sadly misses the odd one, but I have 'skip30' and 'back5' buttons on my remote to solve that - 7 or 8 quick clicks past the ads, then back to the start of the prog.
I haven't seen an ad in many months. TV has improved out of sight for me.
IAADBD (I am a DB designer) who has designed and implemented DBs for large apps (millions of rows, hundreds of entities). If your data is important, you need to look at this from the data's point of view.
I achieve these by permitting access only via stored procedures (with rare, simple exceptions), and having complete coverage with DRI (declaritive referential integrity), triggers, defaults, rules, timestamp checking and so on. I control what access people/processes have to the data, and am able to encapsulate people away from the underlying structures.
I can therefore be sure the data stays clean and well formed. I can also be sure that people are not pulling large gobs of data out of the database and across the network to implement business logic that should be done inside the DB.
If your data can be adequately protected using DRI alone, and you are not so concerned with encapsulation and performance then these tools are brilliant.
Horses for courses, as always.
Put the water cooler and pump outside.
Then you can seal the box. The water cooler
might get dusty, but the PC wont. Which do you
care more about?
So these RDBMSs provide tools - triggers, stored procedures, a multitude of locking mechanisms, DRI (Declaritive Referential Integrity) and so on. With a well built database:
- You *can't* insert duplicate data
- You *never* modify data behind a users back (e.g. remove trailing spaces on varchar)
- You *never* break a foreign key, leaving dangling data
- You can perform all maintenance while the DB is live
- and so on...
The SQL standard is very clear and explicit about data handling - in general no surprises and never let in bad data. And all 'industrial strength' RDBMSs do a good job of following the standard... sadly not perfect but very good. MySQL is not one of them - see here for some of the ugliness inside MySQL exposed. For example: MySQL's use of NULL is just bizarre, as is its ability to accept invalid data and change it silently.'Industrial Strength' RDBMSs are about predicability and reliability. Speed is important, correctness is essential. MySQL aint there yet. Oracle, MSSQL, Postgres, Sybase, to name a few of my favourites, are.
Let the flames fly :-)
Protests need publicity. All protests are publicity stunts.
If you had some libraries that you wrote in your time and brought it, you will want to keep them.
Actually (small correction) - I have an 80W power supply. I have a full sized hard drive, DVD burner and an encoder/decoder card (Hauppauge PVR-350), external HDD via firewire. With the 60W and everything loaded, I had intermittent failures on start. When I put the 80 on, I found that it drew about 72W very briefly after start. Its now reliable.
Did you see the power supply - 60W - lets see you find a shuttle that runs on that!
I have one of these (MII12000 in fact) - idle draw is less than 20W - a dim bulb.
Power efficient, cool, quiet, small, but not very fast.
Pick the right tool for the job - in my case a PVR.
They will find him.
Eventually he will step over some line and annoy SteveB. At that point, MS will sue and subpoena logs from blogspot. IP address and personal details will quickly be revealed.
Kinda makes you wish for true anonymity in blogging that a tool like freenet could provide (if it weren't such a dog slow pile of dung).
The fact that this is such a fast moving standard is very interesting. Most standards take years of negotiation to move forward - with MS at the helm this one is fair trucking along. In effect, this makes the standard a moot point - it is unlikely that anyone will have the resources or willpower to create a competitive implementation to C# given the goal posts move so rapidly. MS have brilliantly created a standard, and probably what will be the only real-world usable implementation of that language. They get the best of both worlds - standards and proprietary. Cool.
Even if the quality is not the best, those of us who pay up to several dollars a minute for cellphone use will find it pretty useful.
I am roaming with GSM - and am thinking of direct crediting my pay straight to vodaphone to save on bank charges. Its just ridiculous - every call (even local to where I am) is charged as a toll call.
I can only imagine mikejz84 is an American who does not travel overseas very much. Local calls in the US are very cheap, wireless plans often build in hundreds (or thousands) of minutes per month. Once you leave Oz, its not such a pretty story.
Also, in the last two years the linux kernel has been spiraling out of control.
.7, 2.6.12.xxx, 2.6.13 just out and so on.
In the past we had the 2. stable branch. Stable, bug fixes, few new features. Then we had the wild west of the 2, branch with a pell-mell of features, bugs and all that good stuff that goes into a fast moving product.
Now we just have the 2.6 tree. Its got all the good stuff but its also the dev branch. Its moving like crazy, stuff is breaking (USB) and being fixed at a great rate.
If you are building a major server, you want a stable platform. You dont want kernel updates every few weeks 2.6.11.1,
I think the move away from odd + even branches has been a net loss to the reputation of the linux kernel.
Another problem is that satellites are affected by numerous gravitational sources (e.g. moon, sun) and slowely wonder off course. For example, equatorial satellites require small adjustments every year. While their mass is very low - there didn't look to be a lot of room inside that box for some thrusters.
My point is that the 2.6 kernel is not without its flaws - it depends which particular kernel version you get and which warts it has.
My advice is if it ain't broke...
Do you really *need* the hassle of the upgrade?
This is a complete waste of time and does not merit the front page of slashdot. C'mon - did Zonk even look at TFA?
Just off the top of my head, there is no:
- comparison of file sizes
- analysis version tracking
- comparison of printing/preview capability
- review of scripting capabilities and availability of scripts
- review of the style system
- interoperability of: templates, objects etc
I am underwhelmed.Given the choice between incompetency and conspiracy (esp with Govt depts) - I choose incompetency every time.