The problem isn't proprietary vs. open source drivers. The problem is, quite simply, a very bad decision made by Linus Torvalds: the decision to have driver ABI's that change from release to release.
All Linux needs is to have the driver ABI's remain stable throughout each major kernel version (2.4.x, 2.6.x, etc.). If the kernel developers want to change the parameters or whatever, they've got to wait for 2.8.0.
This would allow each user, each developer, each organization that works with Linux to make the decision on their own as to whether they want to use and/or develop closed source drivers. Sure, in a Stallman utopia, everything would be open source all the time. But the reality is that that utopia is still a work in progress, and most of us would be quite happy to just download the latest driver for an NVidia graphics card or an Intel wireless chipset, plug it in, and let it run. If Linux is to gain mainstream acceptance at the desktop level, drivers have to "just work." And that's not going to happen when you expect users to compile their own drivers.
Think outside the box. Calendars, email, etc. are useful tools, but it'd be a mistake to try to do everything exactly the way Microsoft does. For an alternative approach, you might want to try Citadel instead. Open source, AJAX-enabled, and talks to lots of client packages out there. And it doesn't try to be a clone of Microsoft's offering -- instead, it starts with the approach of helping your user community to work/play/quack *as* a user community instead of as a bunch of disparate users who happen to be occasionally sending data back and forth. Give it a try.
Email is popular because it gets the job done. People like to have a "flow of conversation" just like they do when talking to each other in person.
This ought to be a lesson to people building collaboration software. Microsoft has a lot of people convinced that calendars and address books are the killer apps for collaboration, but in reality, people are looking to be connected to other people. I may be a little bit biased on this one, though, because I'm involved in a project that has built a collaboration software with its roots in the BBS world. Our user community loves it because it lets people work/play/quack together without calling attention to itself.
Funny how they keep bringing up deep-pocketed Google as the alleged "free rider" (funny, I thought Google
paid their ISP bills just like the rest of us) but Google is a straw man. This is just another blatant
attempt by the incumbent telecom scumbags to "tax" Vonage and Skype into price parity with conventional
telephone service.
Even though us geeks tend to see little value in having a calendar bolted to an email program, there are lots of people out there who just can't seem to live without it. So this is a good first step.
But don't go looking for the one big server app that's going to be the "Exchange Killer" that goes with it. That's not how the open source world is answering that challenge. Exchange will not be a Goliath felled by David, it will be more like a Gulliver restrained by multiple Lilliputians. This is because programs like Lightning aren't being written to work with a single server -- they're using Webcal (iCalendar publish/subscribe over HTTP, made popular by Apple of course) and can talk to groupware servers like Citadel and OpenGroupware today. Further on down the line, connectors will become available for the emerging standard GroupDAV protocol. For more complex server-side logic, eventually CalDAV will come out of draft as well.
It's going to be a great world. Finally, after all these years of delay, group calendaring and scheduling will be as open, interoperable, and non-dominated-by-one-player as email is today.
The more the various open apps work together with the same open protocols, the more easily the Lilliputians will surround the hapless Gulliver, and take him (Microsoft) down to their level.
You're absolutely right. There will be no "Exchange killer" because there are lots of open projects which operate in that space. The standard protocols and formats are, as you pointed out, beginning to converge.
I'm involved in the development team for Citadel, another open source groupware platform. We have a friendly relationship with most of the other projects that operate in this space, and that's a good thing, because collaboration on standard formats and protocols is exactly what we all need to take on Microsoft as a group.
(And a big "STFU" to those idiots at Novell who claimed last year that no one is in this space and they're going to take it over. I hope they're the first against the wall when the revolution comes.)
keeping them too busy to crush the smaller players entering the groupware market
Microsoft happened to score big with Outlook and Exchange because they bolted a calendar and an email program together at the right time. They know that this is a big lock-in point for them. Unfortunately for Microsoft, end-to-end support for integration of the most popular groupware features (email, address book, calendar, tasks, and notes) are rapidly coming together in open source offerings, so they've been trying to create the next big lock-in opportunity by "collaboration enabling" all of their software. They want to get their customers into a state where they need to have Office bolted to SharePoint in order to do anything useful. The problem is, most users don't care. They don't need to have spreadsheets that are active through three desktops at once. They just want the damn calendar to work. This gives the open source offerings some time to catch up.
Anyone who goes looking for the Next Big Thing (tm) isn't going to find it. It's not predictable. We never know whether something is going to catch on in a big way, until after it happens. All you can do is sit back and wait to see what people are paying attention to.
Take a look at Microsoft, for example. They have a huge war chest full of monopoly money and they have been actively trying to create the Next Big Thing for nearly two decades now, and not once have they succeeded. Don't you think that if it were possible to predict the Next Big Thing, that those with the financial and political means to do Whatever They Want (tm) would have a virtual lock on it?
In technology, the innovations that change everything come from where you least expect them. That's because the big dogs have a vested interest in preserving the status quo.
Novell has a long history of not being able to get out of its own way. From the bumbling days of DR-DOS after they acquired it from Digital Research, to the "Univel SuperNOS" project (brilliant idea -- they were going to fuse together Netware and Unix -- and they simply abandoned the project and let Microsoft eat their market), to their latest move of letting Ximian take over SuSE (let's be realistic here, that's how it ended up happening), there just doesn't appear to be anyone over there who knows how to actually execute a plan and drive technology into marketable products.
So what's going to happen? My guess would be that the new technology like Xgl and its associated compositing managers etc. will find their way into the pool of open source software, and then someone like Ubuntu will drive it into a slick, easy to install, easy to use desktop that people will actually want to use.
And then when Novell's revenues continue to slide, even these technologies will lose their staffing, when the next round of layoffs will cut those who are not working on products and services that directly generate revenue. Seriously, the whole Novell organization ought to be divested and sold off as pieces to other companies who can work with the products and services that still have some value.
I see maybe three or four names in that list. Certainly not the full-time team of 25 or more that they were touting when the project was announced.
There's really a lot of bad blood associated with this project. They exploded onto the scene, declared that nothing exists in this space, and that they're going to take it over. This was very insulting to all of the other existing open source groupware projects, most of which have a friendly relationship with each other. As a result, most of us were happy to hear about the layoffs.
Hence the XBox 360's built-in Media Center Extender. Leave the media PC in the den; access the content from game console in the Living Room.
That still doesn't solve the problem. You're still futzing with a PC (complete with all of the problems of a PC) even if it is not physically in your A/V rack.
Is it just me, or has Microsoft been pushing Media Center really really hard lately? Mainly through box makers like Gateway and Dell? It seems that none of their strategies to monopolize the living room seem to be panning out, so now they're just doing a Dresden-style bombing of the market, pushing harder and harder and louder and louder until someone out there eventually decides to buy Media Center.
The bottom line is that most consumers just don't want a computer in their living room. They want consumer electronics that "just work," like TV's and VCR's and DVD players and surround sound amplifiers. At the end of the day when they plop down in front of the tube, they don't want to have to contend with worms and viruses and email and crashes and software installation/uninstallation and all of the other headaches that go with a typical PC (the availability of better OS's notwithstanding) -- they just want to switch it on and veg out!
The rant you are referring to sounds great at the surface, but it only goes to prove that JWZ does not have much of an understanding of what "groupware" is. Sure, the workflow and process management stuff he was talking about is very boring and only interesting to middle managers, but groupware covers so much more than that. Any system that lets users connect and collaborate, that lets them do what they do best as a team or as a community, is groupware. Calendar apps are groupware. Message boards (yes, even Slashdot) are groupware. MySpace is groupware. Groupware is GOOD, and it's thriving more than ever now that everyone is so well connected!
By the way, the Hula project is pretty much dead at this point. The dev team was sacked as part of the big Novell layoff last November. Dave Camp (formerly of Ximian) is kind of limping it along at this point, so perhaps some interesting bits of code may trickle out in the distant future, but don't expect this project to bear fruit anytime soon. Not that it matters -- even though they claimed that they were "filling an unclaimed spot" in the free software world, there are half a dozen other collaboration platforms out there that solve the same problems admirably. Go check them out.
The 1981 breakup gave the Baby Bells "local dialtone" and big AT&T (and others) "long distance." Now that the technology has all changed, this line of demarcation is obsolete.
It's time for another breakup, and this time it should go as follows: the RBOC's (soon to be the One Big BOC) maintain the physical cable plant, and they maintain the central offices basically as colocation facilities. Then, you have carriers (none of which are allowed to be RBOC's [or the imminent One Big BOC]) as colocation customers in those central offices. They lease customer loops from the BOC/LEC/whatever and then they provide "telecom services" over those loops. We don't care what the services are -- dial tone, DSL, whatever. No distinction between voice and data, between local and long distance, whatever, because as we know, it's all the same crap now.
THAT is the perfect way to keep the government-granted monopoly working efficiently for consumers. The monopoly must extend only as far as it needs to, and no further.
Edison and Tesla were both right. Remember, the DC vs. AC wars were fought back when the load was mostly made up of lights, motors, very utilitarian things. AC is fantastic for transmission over long distances (and for running three phase motors, but that's another story). DC happens to be better at running precision equipment like computers -- heck, they all run on DC already. All we're really talking about here is taking advantage of an economy of scale by doing one big power supply (or a few, for redundancy) instead of one for each machine.
Ever seen a telco rack? Everything runs on -48VDC. Everything. A telco rack always includes a couple of DC power supplies, and all the equipment just ties in to a common DC bus. The best part of all: the UPS simply consists of four "car batteries" (not exactly, but you get the idea) wired in series and tied directly into the bus! No pesky inverters to deal with.
The telecom industry has been doing it this way for decades. It's about time the computer industry got on board.
We can learn from them, too. For example, everyone thinks that our project's FAQ is far more professional and business-like now that we've changed it from a "FAQ" to a "Knowledge Base."
Disregard this completely. Networks will add capacity to meet the demand of their paying customers.
If you ask me, this is probably a bunch of scaremongering by the pigopolists over at Verizon and AT&T designed to get people to think more highly of the idea of a "tiered Internet" where "content providers" like Google have to pay extra for the privilege of sending bandwidth-intensive video over the Internet.
I get the impression that Oracle is just doing this to screw with MySQL. As many know, MySQL gives you a choice of back end data stores. You can go with MAX (now owned by Oracle), or you can go with Berkeley DB (now owned by Oracle).
As the developer of an application that uses Berkeley DB for all of its data stores, I am more than a little concerned about this. Does Oracle see any actual value in Sleepycat, or are they just doing this to shut them down?
Naturally. MS would love to charge everyone a monthly fee for each game, and shut it off when you stop paying for it. It's quite sad that in 2030, you might still be able to come across a box in the attic containing an Atari 2600 and some games, and still be able to play them, but if you come across a box with an Xbox 3 (or whatever) it'll be a useless hunk of plastic.
I've been looking at the free calendaring disaster for a while now - and it is; there are
perhaps 5-10 different packages, none of which interoperate; some very nice clients that
only talk to really crap servers and some very nice servers with poor clients.
Funny, that's exactly the problem that GroupDAV is supposed to solve. On the client side, Kontact and a few others support it; on the server side, Citadel and OpenGroupware.org (plus a few others) support it. It's just a matter of getting the other major players on board.
It's actually not a "telnet" UI per se, but simply a text mode user interface. If you don't like it, then all you have to do is... not use it! Keep in mind that Citadel works in a lot of different use cases. The text mode UI is really intended for BBS applications, so if you're not running a message board on your Citadel server then it's not really the client for you. You'd be better served running WebCit (the AJAX-enabled web interface) or even some of the fat clients such as Kontact that are well-integrated with Citadel.
Funny, it seems that the idea behind Hula is really a complete ripoff of the idea behind Citadel, and the Citadel team has been pushing this concept since 1998 or so. Moving from corporate-focused groupware towards the modern idea of social software isn't something you can retrofit; it has to be the heart and soul of the design.
Hula only exists because some immature but influential people at Ximian (who call themselves 'Novell' these days) have a serious case of Not Invented Here syndrome.
Google is not turning itself into a new version of AOL or Compuserve. Google is, however, quietly building out its own network infrastructure. Right now anyone who wants to can do BGP peering with Google at any NAP it happens to have built out to. What does this buy them?
Let's say that I'm a mid size ISP (I happen to work for one so this is a first hand account) and I peer with Google at a regional NAP. What happens then? Any traffic between my network and Google's network will cross that peering point. As a result, I don't have to pay one of my upstream ISP's for bandwidth to Google. Google, in turn, doesn't have to pay their upstream ISP's for bandwidth to my customers. Everyone wins (except for the upstream ISP's of course).
Any large network operator is already doing this kind of thing on a large scale. Google is already doing this. The reason they bought all of that dark fiber is so they can do it without having to rent a bunch of OC-48's from the phone company in order to make it happen. There is no secret, so stop trying to figure it all out.
Inevitably there will be operations in place in which some company goes in and offers to trade laptops for something they really want (such as... I don't know, food, maybe?) and then sells the $100 laptops back to customers in nations that can afford computers.
I'm looking forward to picking up one of these cheapie laptops in a year or two, even if it ends up costing $200.
Slashdot Mirror
The problem isn't proprietary vs. open source drivers. The problem is, quite simply, a very bad decision made by Linus Torvalds: the decision to have driver ABI's that change from release to release.
All Linux needs is to have the driver ABI's remain stable throughout each major kernel version (2.4.x, 2.6.x, etc.). If the kernel developers want to change the parameters or whatever, they've got to wait for 2.8.0.
This would allow each user, each developer, each organization that works with Linux to make the decision on their own as to whether they want to use and/or develop closed source drivers. Sure, in a Stallman utopia, everything would be open source all the time. But the reality is that that utopia is still a work in progress, and most of us would be quite happy to just download the latest driver for an NVidia graphics card or an Intel wireless chipset, plug it in, and let it run. If Linux is to gain mainstream acceptance at the desktop level, drivers have to "just work." And that's not going to happen when you expect users to compile their own drivers.
Think outside the box. Calendars, email, etc. are useful tools, but it'd be a mistake to try to do everything exactly the way Microsoft does. For an alternative approach, you might want to try Citadel instead. Open source, AJAX-enabled, and talks to lots of client packages out there. And it doesn't try to be a clone of Microsoft's offering -- instead, it starts with the approach of helping your user community to work/play/quack *as* a user community instead of as a bunch of disparate users who happen to be occasionally sending data back and forth. Give it a try.
Change the DNS to make GPS.dix.dk a CNAME that points to pool.ntp.org, and then put the stratum-1 server somewhere else.
Email is popular because it gets the job done. People like to have a "flow of conversation" just like they do when talking to each other in person.
This ought to be a lesson to people building collaboration software. Microsoft has a lot of people convinced that calendars and address books are the killer apps for collaboration, but in reality, people are looking to be connected to other people. I may be a little bit biased on this one, though, because I'm involved in a project that has built a collaboration software with its roots in the BBS world. Our user community loves it because it lets people work/play/quack together without calling attention to itself.
Funny how they keep bringing up deep-pocketed Google as the alleged "free rider" (funny, I thought Google paid their ISP bills just like the rest of us) but Google is a straw man. This is just another blatant attempt by the incumbent telecom scumbags to "tax" Vonage and Skype into price parity with conventional telephone service.
Even though us geeks tend to see little value in having a calendar bolted to an email program, there are lots of people out there who just can't seem to live without it. So this is a good first step.
But don't go looking for the one big server app that's going to be the "Exchange Killer" that goes with it. That's not how the open source world is answering that challenge. Exchange will not be a Goliath felled by David, it will be more like a Gulliver restrained by multiple Lilliputians. This is because programs like Lightning aren't being written to work with a single server -- they're using Webcal (iCalendar publish/subscribe over HTTP, made popular by Apple of course) and can talk to groupware servers like Citadel and OpenGroupware today. Further on down the line, connectors will become available for the emerging standard GroupDAV protocol. For more complex server-side logic, eventually CalDAV will come out of draft as well.
It's going to be a great world. Finally, after all these years of delay, group calendaring and scheduling will be as open, interoperable, and non-dominated-by-one-player as email is today.
I'm involved in the development team for Citadel, another open source groupware platform. We have a friendly relationship with most of the other projects that operate in this space, and that's a good thing, because collaboration on standard formats and protocols is exactly what we all need to take on Microsoft as a group.
(And a big "STFU" to those idiots at Novell who claimed last year that no one is in this space and they're going to take it over. I hope they're the first against the wall when the revolution comes.)
Anyone who goes looking for the Next Big Thing (tm) isn't going to find it. It's not predictable. We never know whether something is going to catch on in a big way, until after it happens. All you can do is sit back and wait to see what people are paying attention to.
Take a look at Microsoft, for example. They have a huge war chest full of monopoly money and they have been actively trying to create the Next Big Thing for nearly two decades now, and not once have they succeeded. Don't you think that if it were possible to predict the Next Big Thing, that those with the financial and political means to do Whatever They Want (tm) would have a virtual lock on it?
In technology, the innovations that change everything come from where you least expect them. That's because the big dogs have a vested interest in preserving the status quo.
Novell has a long history of not being able to get out of its own way. From the bumbling days of DR-DOS after they acquired it from Digital Research, to the "Univel SuperNOS" project (brilliant idea -- they were going to fuse together Netware and Unix -- and they simply abandoned the project and let Microsoft eat their market), to their latest move of letting Ximian take over SuSE (let's be realistic here, that's how it ended up happening), there just doesn't appear to be anyone over there who knows how to actually execute a plan and drive technology into marketable products.
So what's going to happen? My guess would be that the new technology like Xgl and its associated compositing managers etc. will find their way into the pool of open source software, and then someone like Ubuntu will drive it into a slick, easy to install, easy to use desktop that people will actually want to use.
And then when Novell's revenues continue to slide, even these technologies will lose their staffing, when the next round of layoffs will cut those who are not working on products and services that directly generate revenue. Seriously, the whole Novell organization ought to be divested and sold off as pieces to other companies who can work with the products and services that still have some value.
I see maybe three or four names in that list. Certainly not the full-time team of 25 or more that they were touting when the project was announced.
There's really a lot of bad blood associated with this project. They exploded onto the scene, declared that nothing exists in this space, and that they're going to take it over. This was very insulting to all of the other existing open source groupware projects, most of which have a friendly relationship with each other. As a result, most of us were happy to hear about the layoffs.
Is it just me, or has Microsoft been pushing Media Center really really hard lately? Mainly through box makers like Gateway and Dell? It seems that none of their strategies to monopolize the living room seem to be panning out, so now they're just doing a Dresden-style bombing of the market, pushing harder and harder and louder and louder until someone out there eventually decides to buy Media Center.
The bottom line is that most consumers just don't want a computer in their living room. They want consumer electronics that "just work," like TV's and VCR's and DVD players and surround sound amplifiers. At the end of the day when they plop down in front of the tube, they don't want to have to contend with worms and viruses and email and crashes and software installation/uninstallation and all of the other headaches that go with a typical PC (the availability of better OS's notwithstanding) -- they just want to switch it on and veg out!
The rant you are referring to sounds great at the surface, but it only goes to prove that JWZ does not have much of an understanding of what "groupware" is. Sure, the workflow and process management stuff he was talking about is very boring and only interesting to middle managers, but groupware covers so much more than that. Any system that lets users connect and collaborate, that lets them do what they do best as a team or as a community, is groupware. Calendar apps are groupware. Message boards (yes, even Slashdot) are groupware. MySpace is groupware. Groupware is GOOD, and it's thriving more than ever now that everyone is so well connected!
By the way, the Hula project is pretty much dead at this point. The dev team was sacked as part of the big Novell layoff last November. Dave Camp (formerly of Ximian) is kind of limping it along at this point, so perhaps some interesting bits of code may trickle out in the distant future, but don't expect this project to bear fruit anytime soon. Not that it matters -- even though they claimed that they were "filling an unclaimed spot" in the free software world, there are half a dozen other collaboration platforms out there that solve the same problems admirably. Go check them out.
The 1981 breakup gave the Baby Bells "local dialtone" and big AT&T (and others) "long distance." Now that the technology has all changed, this line of demarcation is obsolete.
It's time for another breakup, and this time it should go as follows: the RBOC's (soon to be the One Big BOC) maintain the physical cable plant, and they maintain the central offices basically as colocation facilities. Then, you have carriers (none of which are allowed to be RBOC's [or the imminent One Big BOC]) as colocation customers in those central offices. They lease customer loops from the BOC/LEC/whatever and then they provide "telecom services" over those loops. We don't care what the services are -- dial tone, DSL, whatever. No distinction between voice and data, between local and long distance, whatever, because as we know, it's all the same crap now.
THAT is the perfect way to keep the government-granted monopoly working efficiently for consumers. The monopoly must extend only as far as it needs to, and no further.
Edison and Tesla were both right. Remember, the DC vs. AC wars were fought back when the load was mostly made up of lights, motors, very utilitarian things. AC is fantastic for transmission over long distances (and for running three phase motors, but that's another story). DC happens to be better at running precision equipment like computers -- heck, they all run on DC already. All we're really talking about here is taking advantage of an economy of scale by doing one big power supply (or a few, for redundancy) instead of one for each machine.
Ever seen a telco rack? Everything runs on -48VDC. Everything. A telco rack always includes a couple of DC power supplies, and all the equipment just ties in to a common DC bus. The best part of all: the UPS simply consists of four "car batteries" (not exactly, but you get the idea) wired in series and tied directly into the bus! No pesky inverters to deal with.
The telecom industry has been doing it this way for decades. It's about time the computer industry got on board.
We can learn from them, too. For example, everyone thinks that our project's FAQ is far more professional and business-like now that we've changed it from a "FAQ" to a "Knowledge Base."
(It's funny. Laugh.)
Disregard this completely. Networks will add capacity to meet the demand of their paying customers.
If you ask me, this is probably a bunch of scaremongering by the pigopolists over at Verizon and AT&T designed to get people to think more highly of the idea of a "tiered Internet" where "content providers" like Google have to pay extra for the privilege of sending bandwidth-intensive video over the Internet.
I get the impression that Oracle is just doing this to screw with MySQL. As many know, MySQL gives you a choice of back end data stores. You can go with MAX (now owned by Oracle), or you can go with Berkeley DB (now owned by Oracle).
As the developer of an application that uses Berkeley DB for all of its data stores, I am more than a little concerned about this. Does Oracle see any actual value in Sleepycat, or are they just doing this to shut them down?
Naturally. MS would love to charge everyone a monthly fee for each game, and shut it off when you stop paying for it. It's quite sad that in 2030, you might still be able to come across a box in the attic containing an Atari 2600 and some games, and still be able to play them, but if you come across a box with an Xbox 3 (or whatever) it'll be a useless hunk of plastic.
I've been looking at the free calendaring disaster for a while now - and it is; there are perhaps 5-10 different packages, none of which interoperate; some very nice clients that only talk to really crap servers and some very nice servers with poor clients.
Funny, that's exactly the problem that GroupDAV is supposed to solve. On the client side, Kontact and a few others support it; on the server side, Citadel and OpenGroupware.org (plus a few others) support it. It's just a matter of getting the other major players on board.
It's actually not a "telnet" UI per se, but simply a text mode user interface. If you don't like it, then all you have to do is ... not use it! Keep in mind that Citadel works in a lot of different use cases. The text mode UI is really intended for BBS applications, so if you're not running a message board on your Citadel server then it's not really the client for you. You'd be better served running WebCit (the AJAX-enabled web interface) or even some of the fat clients such as Kontact that are well-integrated with Citadel.
Funny, it seems that the idea behind Hula is really a complete ripoff of the idea behind Citadel, and the Citadel team has been pushing this concept since 1998 or so. Moving from corporate-focused groupware towards the modern idea of social software isn't something you can retrofit; it has to be the heart and soul of the design.
Hula only exists because some immature but influential people at Ximian (who call themselves 'Novell' these days) have a serious case of Not Invented Here syndrome.
Google is not turning itself into a new version of AOL or Compuserve. Google is, however, quietly building out its own network infrastructure. Right now anyone who wants to can do BGP peering with Google at any NAP it happens to have built out to. What does this buy them?
Let's say that I'm a mid size ISP (I happen to work for one so this is a first hand account) and I peer with Google at a regional NAP. What happens then? Any traffic between my network and Google's network will cross that peering point. As a result, I don't have to pay one of my upstream ISP's for bandwidth to Google. Google, in turn, doesn't have to pay their upstream ISP's for bandwidth to my customers. Everyone wins (except for the upstream ISP's of course).
Any large network operator is already doing this kind of thing on a large scale. Google is already doing this. The reason they bought all of that dark fiber is so they can do it without having to rent a bunch of OC-48's from the phone company in order to make it happen. There is no secret, so stop trying to figure it all out.
Inevitably there will be operations in place in which some company goes in and offers to trade laptops for something they really want (such as... I don't know, food, maybe?) and then sells the $100 laptops back to customers in nations that can afford computers.
I'm looking forward to picking up one of these cheapie laptops in a year or two, even if it ends up costing $200.