Slashdot Mirror

I don`t know if there is a version of windows with support for more than 256 logical processors (whatever that means). http://www.microsoft.com/windowsserver2008/en/us/r2-scalability-reliability.aspx

http://www.microsoft.com/windows/pc-scout/

This is a nice tool. Enter the parameters you want, and there's a list of suggestions.

Try:
http://www.microsoft.com/windows/pc-scout/

From their site:

Our interactive guide will recommend just the right PC. We'll fill you in on laptop basics, and match your needs against a pool of PCs recommended by the experts here at Windows.

You can use the 'Get Started' link to begin a guided selection or use the Icons below that to select a laptop type.
This will give you a much better idea of of what (you and) your wife are looking for along with prices for your region/area :)

    Well, I could describe it in detail, but Microsoft has been nice enough to have plenty of step by step instructions.

http://msdn.microsoft.com/en-us/library/879kf95c.aspx

    These are the step by step instructions on how to "write" a web site, with user registration, authentication, management, public and members only pages. I quoted "write", because you don't write one line of code. The only typing is to change the default filename to what you want, to enter your username and password, and to put in whatever text you want.

    People who do this consider themselves "developers". If you buy VS2010, you'll be spending from $799 to $2,169 (MS MSRP). Then you need a Windows server to go with it. That's $469 to $2,999 (MS MSRP). Oh and the DB. MSSQL server is $3,500 to $54,990. Since you've gone that far, you'll want your Exchange server to go with it too, for $699 to $3,999. I didn't even go into the CAL's, And lets not forget, one set of servers isn't safe, you want redundancy. And you want a couple Active Directory servers, and the matching Microsoft Professional (or higher) workstations so they can join the domain, and Microsoft Outlook (almost always purchased in the Microsoft Office package).

    In the end, you have a developer who you've paid a lot of money to, a whole lot of money spent on Microsoft software, and servers. And what do you have? An enterprise based on something that someone pointed and clicked through, but they can't give you extra functionality. If you want something outside of the scope of the warm fuzzy point and click environment, you'll spend an extra fortune, and still be vendor locked and screwed when the platform is obsolete. Sure, it's great in 2011, that you've built everything in on their 2008 to 2010 platform, but what happens in 2016, just 5 years from now? Take the whole exercise, and do it again, so your developer (or another one who sings the praises of Microsoft) can point and click through another "enterprise" web application that your enterprise will be shoe-horned into. It may or may not suit all of your needs, but since you spent so much money on it already, it must be the "right" way. And sure as hell now you can't just switch to another platform.

    Or you go with a real programmer, who can sit down and write something in C/C++, Perl, PHP, Python, Ruby, etc, etc, etc... If they do their jobs properly, they'll have the job done right, on time, and way under the budget outlined above.

    Did I flesh that out enough for you?

That has a link to the exact terms, with examples.

I think you have to do a software as a service but you don't have to host with MS. They will push it hard but you can host where ever you want.

I completely agree with you about everything -- particularly about developers who suffer from extreme 'not-coded-here' syndrome -- except the last line:

I'd say startups don't use .NET and Windows in general, because of licensing. Simple. They don't have to cash to do it. You might also find that the people who have worked at startups are used to dealing with this, because of their own monetary constraints.

I'd argue that a medium size startup isn't going to care about the cost of software licencing. It's a drop in the bucket compared to office space, computer hardware, office equipment, and personnel.

For a small developer start-up, I can't see them not using the MS Action Pack. It's free to be a MS Partner, and then $300/year to subscribe to the Action Pack. That gets you (among other things):
Office 2010 Pro Plus x10
Project Pro 2010 x5
Visio Pro 2010 x10
Exchange Std 2010 /w 10 CALs
SQL Server 2008 Enterprise R2 /w 10 CALs
Windows 7 Pro upgrade x10
Server 2008 R2 Ent /w 10 CALs
Web Server 2008 R2
SBS 2008 R2 /w 10 CALs

All the above are real SKUs of the software. Not developer versions. But if you want developer versions, you get those, too, with the 3 MSDN Subcriptions. The three subscriptions alone would typically run you $2,000 apiece.

https://partner.microsoft.com/40016455

Are you a small start up? Go out and buy your barebones servers, get half a dozen consumer-grade laptops, and get the Action Pack. Done.

Actually, the reason why size() is O(N) in gcc is because, if you have a counter in the list itself, then splicing the tail of another list into yours becomes an O(n) operation (because you need to count how many elements are there to splice to update the counter). Thus, you either have O(1) size(), or O(1) splice(), but not both. VC and gcc had different ideas about which one of those operations is more common.

Having the counter itself is hardly a space or performance hog, especially given that std::list is a doubly linked list, but is actually used where a singly linked one would do in the majority of cases in C++ programs (since C++03 didn't offer a singly linked list container; now we hve std::forward_list in C++0x). Keeping those back pointers around and updating them is much more of a hog than a single counter.

Anyway, in my opinion the strongest argument in favor of O(1) size() is that all other containers are O(1). Thus generic code can reliably use it regardless of type of container.

They have another system in place to handle certificate revocation. You can enable/disable this in IE in "Internet Options->Advanced->Security". I believe Safari and Chrome also use this OS level certificate handling too.

This is actually one of the improvements in Windows Vista:

The TaskDialog is the OS functionality for easily showing a dialog with descriptive button labels instead of just old school MessageBox with OK/Cancel/etc.

I'm glad we have our own CA and can self-sign.

As said, these companies have just been lucky enough to be in the right spot at the right time to have their root-CAs int the browsers.
Interestingly, at least Microsoft has a page detailing the requirements for the CA, should it wish to be part of that list:
http://technet.microsoft.com/en-us/library/cc751157.aspx

How does it work for Firefox?

A citation is needed for every comment [citation: see this comment].

Geez, don't any of you know how to use google any more?

Two Screens Are Better Than One.

Of course, once everyone goes dual/triple/quad, Microsoft will then charge extra for "Window MultiMonitor Edition", and FUD-packer Florian Mueller will claim that Microsoft has various patents that prevent linux from using multiple monitors, but that's another story.

http://research.microsoft.com/en-us/news/features/vibe.aspx

The first study revealed that the users' productivity increased by 9 percent. Further studies showed even greater increases - at times up to 50 percent for tasks such as cutting and pasting. Mary Czerwinski, the VIBE research manager, is excited about her group's discoveries, asking, "If you're able to squeeze 10 percent more productivity out, do you know how much money that will save?"

The article is utter garbage.

What you need is this hotfix but since Samba does NOT conform to the Windows indexing protocol it WILL be slower than native.

If the NAS is an x86 or x64 machine I would just pick up a cheap copy of Win 7 Home Server (or real WinServer, look around you can find copies of WinServer SBS usually cheap) and then Win 7 will give you native speed on indexing network shares.

It is no different than how Linux machines don't index Windows shares at native speed, they simply don't speak the same language, but the hotfix will give you a "quick and dirty" workaround albeit one that is slower than native. Finally this link will give you a nice overview and lists which combos are supported.

Sadly as has always been the case with Samba they are usually a day late and a dollar short, even though MSFT released the protocol specs as part of their deal with the EU. Can't really blame them though, as the FOSS insistence on "free as in beer" simply doesn't give them the extra manpower required to keep up with such complex protocols. If past performance is an indication you should have Win 7 native search integrated in Samba about the time Windows 9 comes out. Sorry.

Right now even if I wanted to I couldn't install IE9 because it isn't supported by XP.

Considering the mainstream support life cycle for XP has ended, it's not really suprising. I do find it annoying when developers waste time on deprecated software honestly.

Why do you say that? IE9 was released to the public on 14th March. I don't see anything at http://microsoft.com/ie/ or http://www.beautyoftheweb.com suggesting it is anything other than a final release.

In Vista and 7, yes, it is in the ultimate version and is called bitlocker.

Windows Vista
http://windows.microsoft.com/en-US/windows-vista/products/compare
Windows 7
http://www.microsoft.com/windows/windows-7/compare/default.aspx

In Vista and 7, yes, it is in the ultimate version and is called bitlocker.

Windows Vista
http://windows.microsoft.com/en-US/windows-vista/products/compare
Windows 7
http://www.microsoft.com/windows/windows-7/compare/default.aspx

Excuse me, AC, but the university I work for paid for a product. Is it wrong to expect the supplier to at least test the patch before it goes out with a decent test coverage? After all, I'm not the only one with the problem.

IE is not a product, it is an abbreviation. Internet Explorer is a registered trademark - both the term and the logo: http://www.microsoft.com/About/Legal/EN/US/IntellectualProperty/Trademarks/EN-US.aspx

Oh, sorry, then I was badly mistaken. I had read those old articles where MS claimed it wasn't a trademark, but I didn't notice the articles which said that they settled for $5M.

IE is not a product, it is an abbreviation. Internet Explorer is a registered trademark - both the term and the logo: http://www.microsoft.com/About/Legal/EN/US/IntellectualProperty/Trademarks/EN-US.aspx

You are wrong if you telling me that if I invented "hastforks" that I could not trademark "Hastfork Store" or "Hast Store" because they are descriptive. Or that Kimberly-Clark can't register "Kleenex Store". If Apple can demonstrate that they marketed Apps before others used the term to describe their programs, suites, wares, or whatever for commercial purposes they should prevail in this.

Not according to them: http://www.microsoft.com/about/legal/en/us/IntellectualProperty/Trademarks/EN-US.aspx

Their refusals to take licenses leave us no choice but to bring legal action to defend our innovations and fulfill our responsibility to our customers, partners, and shareholders to safeguard the billions of dollars we invest each year to bring great software products and services to market,” he added.

The patents at issue cover a range of functionality embodied in Android devices that are essential to the user experience, including: natural ways of interacting with devices by tabbing through various screens to find the information they need; surfing the Web more quickly, and interacting with documents and e-books.

This from the company that derided tabbed browsing as something that end users didn't want when Firefox had it and IE didn't.

They're doing a great job protecting their "great software products and services" - they're so protected that nobody's seen them.

This is what you do when your own product (WP7) can't compete on the merits.

Seriously?

• Method 1: Click on the little circle in the bottom left (or hit the "Win" key on your keyboard) and start typing.
• Method 2: Open an explorer window and type in the little box on the far top left, just under the "X". (Or hit ctrl-f) Here's an image in case you're having trouble.

Once you're there you can either use a full-text search, or use filters like "Tags:", "Rating:", "Genre:", or "Artist:" to search though metadata.

Server Name Indication [...] is still not universally available and thus it will still take some time

In this case, "some time" is until April 2014, the announced end of extended support for Windows XP, or two years after Android "Ice Cream" phones start shipping, whichever is later. Neither IE on Windows XP nor the browser on Android 2.x supports SNI.

The brackets in c++ have a specific purpose and aren't redundant. They define "scope". Making them only for if and while statements means that you wouldn't be able to define scope.

http://msdn.microsoft.com/en-us/library/b7kfh662(v=vs.80).aspx

This is a very good point, the overhead of API calls can be a significant bottleneck.

I'd suggest that a good solution is to move applications to entirely managed code (e.g. C#), so that there is no need for any hardware-enforced barrier between the kernel and the applications (c.f. Singularity). In the best case, you may end up with a situation in which a JIT compiler inlines parts of the kernel's graphics driver directly into the application code, effectively run-time specialising the application for the available hardware. We already see hints of this happening, for instance the use of LLVM bit code in Apple's OpenGL stack.

Why use any commercial solution at all? You're in a school after all. Make your students create simple buttons on your electrics class. You can just transfer the clicks via wire. Then make your programming class code the back-end system for a computer. Microsoft offers Visual Studio Express to students for free and you can easily get a cheap license to your school by calling Microsoft's sales team. At the same time your students will get real world experience and get to know the best programming tools used in the industry.

Come on, you're a Microsoft sales rep., aren't you?

Why use any commercial solution at all? You're in a school after all. Make your students create simple buttons on your electrics class. You can just transfer the clicks via wire. Then make your programming class code the back-end system for a computer. Microsoft offers Visual Studio Express to students for free and you can easily get a cheap license to your school by calling Microsoft's sales team. At the same time your students will get real world experience and get to know the best programming tools used in the industry.

The only difference is that most Linux distros will ask you to enter your password and click OK, whilst Windows 7 will display a big yellow-topped box and just ask you if you're sure.

This can be reconfigured to prompt for a password. I have no idea why it isn't by default.

Of course, finding this setting is a pain... you have to run secpol.msc (msc files open Microsoft's Management Console) and find the entry named "User Account Control: Behavior of the elevation prompt for administrators in Admin Approval Mode", then change the setting to "Prompt for credentials" as documented here.

Yet they don't either clean up their operating system's act by actually making it semi-secure or make their own antivirus software, something that really does seem as though it should happen at the deep OS level

By all accounts Ive seen, you are wrong on both points.

Windows 7 is by any standard I could come up with as secure as either MacOSX or classic Linux, if not more so, because it is actually "battle-hardened" by the attention of every malware writer on the planet. It supports all the latest security mechanisms, from sandboxing (IE8/9), ASLR, DEP, kernel patch protection (PatchGuard), system file protection (SFC), non-admin by default (UAC), and all the rest. Possibly you could argue some breeds of Linux with AppArmor and SELinux are more secure, but as a general rule, windows stacks up quite nicely against its competition. Blaming MS for the programs its users authorize, or for the lax security of Adobe, isnt really fair.

As for antivirus, MS cannot prevent competitors from selling AV, and that would in fact likely get them smacked with another antitrust suit. They do, however, make both AV for home users, and AV for big business.

Yet they don't either clean up their operating system's act by actually making it semi-secure or make their own antivirus software, something that really does seem as though it should happen at the deep OS level

By all accounts Ive seen, you are wrong on both points.

Windows 7 is by any standard I could come up with as secure as either MacOSX or classic Linux, if not more so, because it is actually "battle-hardened" by the attention of every malware writer on the planet. It supports all the latest security mechanisms, from sandboxing (IE8/9), ASLR, DEP, kernel patch protection (PatchGuard), system file protection (SFC), non-admin by default (UAC), and all the rest. Possibly you could argue some breeds of Linux with AppArmor and SELinux are more secure, but as a general rule, windows stacks up quite nicely against its competition. Blaming MS for the programs its users authorize, or for the lax security of Adobe, isnt really fair.

As for antivirus, MS cannot prevent competitors from selling AV, and that would in fact likely get them smacked with another antitrust suit. They do, however, make both AV for home users, and AV for big business.

Let's recall this is the company in the 80s which provided vital parts of the DOS "OS" under undocumented function calls, leaving competitors with crippled options (technical or business wise): http://www.htl-steyr.ac.at/~morg/pcinfo/hardware/interrupts/inte8980.htm . I was reverse engineering much code at those times, only to find out that Microsoft was using the undocumented calls, games were using them, any virus around was using them, ONLY competitors were left in cold! It is also the same company which cripples content, on recipient basis: http://support.microsoft.com/kb/95555 . It is also the same company which does not honor even its own standards, with an incredible impact on consumers (and crippling their freedom): http://arstechnica.com/microsoft/news/2010/04/iso-ooxml-convener-microsofts-format-heading-for-failure.ars & http://en.wikipedia.org/wiki/Office_Open_XML -> Compatibility_between_versions . If you don't get this, check well the attachments in your mailbox. Is tampering with government and standards bodies "ethical" "business" "attitude"? I think Ethisphere Institute has no place near any ethical list.

Yeah, good point. I guess none of the managers or executives responsible for that really egregiously unethical shit they pulled, as recently as 2008, are still at Microsoft. Stand up guys like these, who have only just joined MS, are, as we type, turning Microsoft into the fine, ethical company it is today. Remember kids: "It's not unethical if Microsoft does it".

Did you mean to suggest Microsoft is a hardware company?

Regardless of the rest of your post, you should know that Microsoft does have a hardware division. There's also the Zune of course. Software is the biggest part of their business, but if you got rid of it, there would still be a fairly sizeable hardware company left over.

We have a saying for what you are advocating and it is called security through obscurity and if you are gonna go that route you are better off using AROS or Ecomstation or any of the other even more obscure OSes.

Of course that is also ignoring there is a downside to STO, and that is when a bug DOES come along you're fucked just as there were several thousand that got bit by the KDE theme virus, even though it was on a relatively obscure site that required an activity that many don't do.

I would argue that using the latest and greatest, Windows 7 X64 with a good AV such as Avast or Comodo and low rights browser like IE or Chromium based you will be more protected in the long run since the OS is hardened and has much tougher security thanks to DEP, ASLR, low rights mode, and file and registry virtualization. You can even harden it further if you like (I've been using this for months, if you have friend on Windows 7 I would recommend you send them this link) by adding Structured Exception Handling Overwrite Protection (SEHOP) .

So while I'm glad it works for you there is a reason why an OS is obscure and in the case of Linux it is because it is a giant PITA on the desktop. Six month update deathmarch, updates that fix one thing and break three more, drivers that rarely survive updates, the fact that it is pretty much useless without CLI (if you don't believe me disable your access to Bash for a year. I had a couple of Linux advocates try it and they had to re-enable in less than 6 months as they couldn't fix anything without Bash), in short for the vast majority it just doesn't work. After all I can make a 100% secure machine by cutting all the wires and burying it in a hole filled with concrete, but that doesn't make the machine very useful does it?

XP is scheduled to go end of life in 2014. (See http://support.microsoft.com/lifecycle/search/default.aspx?alpha=Windows+XP) Corporate IT is going to have to start upgrading/replacing XP desktops over the next 3 years to Windows 7 (and maybe Windows 8). Win7's first service pack just issued a few weeks ago. So the whole "wait until the first service pack" crowd doesn't have an excuse anymore. (And we'll ignore Vista. It's best to ignore it. Do not speak it's name lest ye offend the computing gods.)

And at this point Windows 7 is probably a better OS that XP - although it does require beefier hardware and definitely more RAM.

So, yes, the majority of corporate systems are still on XP. But that is going to change over the next 2-3 years. Corp IT isn't going to have a choice, eventually, if they still want security updates/patches. And any IT department that ignores the lack of patches for XP after 2014 is utterly negligent. So time it's time to at least have a grand XP to Win7 upgrade/conversion on your radar, like it or not.

So IE9 (and IE10, etc...) might not be relevant for corp IT right now. But it will be. Don't ignore it.

Says Sarver:

For example, some developers display “comment”, “like”, or other terms with tweets instead of “follow, favorite, retweet, reply” - thus changing the core functions of a tweet.

So clearly one of their fears, perhaps for legal reasons, is getting their functions and widgets confused with, say, Facebook's. We all know how often YouTube and Facebook get mentioned in the same breath as Twitter (heed the great prophet Conan) but they are very much separate companies, and if there's even a chance they're stepping on Facebook's or any other's proprietaryGoodness(TM)* with software patents, DMCA, ACTA and such then I'm not sure twttr will want to take it. It also makes me wonder if Amazon.com asked Facebook before they made their own liker.

It all reminds me of, among other things, part of the Games for Windows Technical Requirements, a piece of terminology:

Vibration Gameplay feedback produced by the controller motor. Do not use rumble.

I definitely would not want to step on Nintendo's toes like that.** Last thing I want Mario to do is wear a black suit and hand me a C&D.

*Forgive me; it's what living in a world with a group called "comScore" does to me.

**My username aside, maybe.

Also, this: http://www.microsoft.com/windows/internet-explorer/compare/default.aspx

I don't use FF or Chrome so I honestly ask which one of those are FUD? I've always treated comparison matricies with suspicion of the cherry picked features shown.

Also, this: http://www.microsoft.com/windows/internet-explorer/compare/default.aspx

Last time I really used it was just before .NET came out.

That would be VS6 - released in 1998; more than 12 years ago. A lot has changed since then.

There's "help" on some library functions and applications that wasn't "just the facts", but instead was meaningless marketing hype

There's certainly a bunch of articles like that on MSDN, usually tutorials or "how-to" things, but I've never seen any such in actual class/method/function documentation. When you press F1 when editor cursor is on a symbol, you invariably end up in a dry, technical article documenting what it does.

Then there were libraries that had been phased out, only you didn't learn that on the first bit of documentation you found.

Well yes, it's not like you're going to have "THIS IS DEPRECATED" in big red letters over every single documentation page for a particular library - just like, say, you don't have that for AWT javadocs. But if you go to the title page of a particular technology, and there is a better replacement available, it will tell you.

As for debugging, the old print statement is still king. Yeah, it's nice to have a debugger, with watches, breakpoints, inspection of variables, etc. That's great for some lame business logic or web app, or for quickly finding where a segfault happened, but not enough for complicated algorithms, where data must be displayed in a way that can be comprehended, and subtle errors will be easily missed. Or where the quantity of data is so great, and filtering out whichever parts are irrelevant is itself not trivial. When your display commands in the debugger get so complicated they ought to be checked in to a repository, you may as well just program that logic into the source code.

"Printf debugging" also fails in the scenarios you describe. Even more so when you have asynchrony and/or multithreading involved. VS, meanwhile, has some special debugging features to tackle that.

As well, in a big project, adding or removing a print for debugging can mean a fairly lengthy recompile. Sometimes you have to do it, but using it as a main debugging technique is more often a waste of time on non-trivial projects.

Code blocks appears date now, but it was actually... nice a few years ago. No headaches, things just worked and I like that. Never had any kind of trouble with Eclipse, so I won't comment on that. But when a piece of software requires 2.3GB of HD space for a simple Visual Basic installation (according to http://www.microsoft.com/visualstudio/en-us/products/2010-editions/visual-basic-express ), is there any question about bloat? The whole point here is "visual studio is the best". It isn't, not in my opinion and many others. Is it good? Maybe, I don't like it, but I can see how some might. But stating that something is "the best" should be more than personal preference or ignorance about alternatives.

It looks like microsoft have ripped off your Word Wiggle game too http://ie.microsoft.com/testdrive/Performance/WrigglyWords/Default.html

COUNT(*) includes null values. COUNT(1) counts all rows where 1=1, which would include null rows. On a modern RDMBS, they're synonymous.

"COUNT(*) is somewhat different in that it returns a count of the number of rows retrieved, whether or not they contain NULL values. " See MySQL 5.5 Reference. Or Microsoft SQL Server saying "COUNT(*) returns the number of rows in a specified table without getting rid of duplicates. It counts each row separately. This includes rows that contain null values.". Asktom remarks, for Oracle, (when asked what the difference is) "nothing, they are the same, incur the same amount of work -- do the same thing, take the same amount of resources."

(Apparently in Oracle 7 and below, COUNT(1) had performance advantages over COUNT(*). I have no idea why.)

(I prefer COUNT(*) because it makes it especially clear I'm really not caring what the value is.)

I don't know exactly what the ANSI spec says, but I assure you that MySQL, Oracle and MSSQL all treat them the same.

Single example, there are many others.

Privilege separation in the default configuration

http://support.microsoft.com/kb/255281

versus

http://www.losurs.org/docs/tips/sysadmin/bind-nonroot

for DNS, for instance, resulting in things like: http://www.cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2007-1748 giving you "root" access to the server itself.

Jesus christ, it's just an IDE

An IDE that Apple and it's fanboys made a big noise about being free, unlike another monopolistic competitor which is said not to offer a version of it's IDE for free.

The problem isn't that they're charging, it's that they are charging when they promised not to. It's a bait and switch. However this isn't the first time Apple have "altered the deal" and if you've locked your self in all you can do is "prey they do not alter it any further". It wont be the last time they do it either.

If MS started charging just $0.05 for Telnet the same people defending Apple would be up in arms. It's the level of sheer hypocrisy that annoys me.

h264 doesn't work, you need a low latency codec. Computing the motion compensation between N keyframes means you're introducing N frames of latency.

So you need to transfer still images, encoded in MJPEG or something similar but more advanced. Is it possible?

of course it is! One solution was introduced recently with the windows SP1, the other one is open source and has been available for some years.

doing it from the cloud (i.e. fancy word for the internet) isn't so interesting, the technology sounds so much desirable on the company's lan, then on the home network. But it still is workable over the internet within conditions of bandwith and latency, i.e. you need a home connexion that both qualifies for HDTV over DSL and a good game of counterstrike. Good DSL may do, fiber would be much better. That's why it already exists again, and sold under the name of "OnLive".

The only thing that doesn't support ARM is "big" Windows 7, and this is changing:
http://www.microsoft.com/presspass/press/2011/jan11/01-05socsupport.mspx

Every update to the Java JVM has the potential of opening up an exploit in that sandbox.

For example:
http://www.microsoft.com/security/portal/Threat/Encyclopedia/Entry.aspx?name=Exploit%3AJava%2FCVE-2009-3867.HD&threatid=2147637070

??? Microsoft did this pretty much as soon as the 360 was released. http://www.microsoft.com/hardware/gaming/productdetails.aspx?pid=091

The last time somebody tested these browsers using Microsoft's Fishtank, Firefox 4 Beta won. I wonder who wins the Fishtank test this time.