Howard MD County Library Chooses Koha

More Good News from Joshua and the gang over at LibLime: Howard County Library in Columbia, MD USA, has selected LibLime's Koha ZOOM Appliance solution as the library's next integrated library system. Koha is the first and most advanced open-source integrated library system.

"Koha was less expensive than other ILS vendors, will allow us to use existing hardware, and is operating-system independent. Koha allows us to provide services to our customers and make our own decisions about the software we run on our desktops -- other ILS options required us to deploy software they want. We prefer to use taxpayer dollars for collections and other core services -- such as our hardware and popular classes, seminars, and workshops."


So silly names make one credible?

Ok, now I've got to weigh in. Two seconds is inaccurate. It's more like 3-5 seconds. The way I understand it, Evergreen uses technology similar to instant messaging. Please correct me if I'm wrong. Instant messaging is not always "instant." Sometimes there is a lag time depending on traffic. Then imagine checking 20 or more items out to one patron. Each scan requires this "instant" message. And it's true, clicking "Done" then waiting for the job to be sent to the printer takes a long time. It's not the speed of the printer, but the speed of the message going to the printer. At least that's my understanding. Then look at the line behind the person in front of you. It isn't pretty.

Why is it that if someone criticizes Evergreen, they are called mean-spirited? I have seen no personal attacks in this forum. Any product out there should be peer reviewed so potential customers can weight the pros and cons. I, for one, don't think Evergreen is all that bad, but it still has a long way to go.

You know, even if the cost is $2M, isn't that about $1 per patron per year? I'm ashamed when I think of how much we spend on ILS maintenance instead of books here.

You know, even if the cost is $2M, isn't that about $1 per patron per year? I'm ashamed when I think of how much we spend on ILS maintenance instead of books here.

What do other systems spend per patron in ILS costs annually?

$1 per capita is very high.

I think the original poster was referring to the manner in which the criticism is being delivered. Being called "incestuous" doesn't summon all that pretty an image and the sarcasm of other posts is exceedingly juvenile. Criticize? Sure, but show some basic level of professional respect.

Why is it in a profession that strives to preserve anonymity so many practitioners have a problem with respecting the desire to remain anonymous?

Because there are two situations that are not quite analogous. In the first, you are seeking to protect the freedom of information rights of citizens from encroachment by the government. In the second, you are attempting to protect yourself from your fellow librarians. Anonymous speech has a venerable history in the pro- anti- freedom interactions between libertarians and the censorial. Precedent in the U.S. dates back to at least 1721 or so, or so I understand. Anonymous speech practices are based on very real concerns over being jailed for uttering dissenting viewpoints. However, I would surmise that in this day and age, the practise is much more likely to be subject to abuse. If you are fired for your viewpoints, you can file a wrongful dismissal suit; the discomfiture suffered will be financial hardship; although that can range from having to tighten your belt to loosing your home and having to live in shelters. I think really, most anonymous speakers these days simply don't want to face censure. Engaging in spirited and lively debate requires first and foremost that you be exposed to ideas.

Mind you, it is all too easy for someone to step over the line and go from censure to harassment. Still, harassment is easily enough dealt with. But those of who pay our dues by standing behind what we utter by doing so openly are irked by those who claim anonymity. Mostly, I think, we read their words and decide that they are doing so only because they take umbrage too easily, and not because they have anything to say that is politically relevant but unpopular.

You don't work in a busy library. It should be a fraction of a second. The receipt printing problem occurs because printing doesn't start until the entire checkout session is done.

Pretty common term in discussing a questionable relationship. Get over it.

The limitation isn't the software, it's the reality of how long it takes to circ a book. Open book, position it under the reader (or wand it), beep, close, set aside, take the next book.

As for cost, that's a bargain for what they get: ILS maintenance, server purchasing and maintenance, salaries for folks to maintain said hardware/software, overdue notice printing and mailing, holds notices, and book courier service.

Kind of makes me wonder who "Anonymous" is.

What you fail to understand is that you have to wait for Evergreent to catch up. You can't get scan the next book until Evergreen is ready.

I'm guessing It's been awhile since you worked circulation. 2, 3, 4, or more seconds to checkout a book is unacceptable. You shouldn't have to wait on the system. Cost? Well free does have a very real labor cost having to keep more staff at the circulation desk.

So positive comments must be from jealous ILS vendors? Just because the point is raised that not all in GA are happy with Evergreen it means that evil is at work.

Let me make sure I'm understanding: you're complaining because a checkout transaction takes 2 seconds? I guess we have different definitions of slow. Your library is sure busier than my metro-area library if 2 seconds per book is an issue.

That is a long time to print a receipt, though. Let me guess... dot matrix printer? About 5-10 years old? We used to have those... not fun, I know. If speed is so paramount, why not carve out a little of the budget for a newer printers that print practically instantaneously?

$1 per registered user, not $1 per capita. If PINES sent 2 overdue notices to a patron in a year that's $1...

Amen, KGS. Some of our fellow librarians must be Bionic Woman in disguise if they can open a book, scan the barcode, close it, and move to the next book in less than 2 seconds. Fraction of a second? I wish my local grocery store moved that fast-but on second thought I have a feeling my eggs would be pulverized.

To some of the other posters here: everyone here would agree that 5 seconds to checkout a book is too much. But the fact of the matter is, as some have pointed out, is that it apparently isn't that way throughout PINES and it seems the majority of libraries are scooting along at a good checkout rate and are happy. As a previous poster said, if checkouts were going at the 5 second clip and things were so "bad and slow"-why would more systems join PINES? I just checked, and it looks like 3 more library systems have joined PINES since they installed Evergreen.

Consequently, that leaves the reader to assume the posters complaining about checkout speed are in one of two camps:

1) they really are having issues, but it's clear Evergreen itself is not the culprit. However, blaming the problems on Evergreen causes a bigger splash than saying, for example, the network is clogged up and causing slow checkout times. Have you even looked at the usage of your internet connection? Right... probably not... because the problem is Evergreen.

2) they are not who they say they are, but rather are people who either fear Evergreen or have some sort of grudge against it. Some of the posters here, with their flamebait one-liners, sound like that.

To the folks in the #1 camp: Look, I have worked (and still often work) the checkout desk, and I know what it's like to have angry customers in line and dealing with software. Evergreen is the easy and visible target here but it's clearly not the right one. The last thing any of us should be doing is taking cheap shots at one of the few good things happening in our profession. Criticism is great, and if you frequent the open-ils lists, you see plenty of it. However, along with those criticisms, you also see positive suggestions to fix those areas that need improving. That's what open source is all about.

I await the "witty" one-liners. Gravy train!

Why would one join PINES? Politically it is a little difficult to ask for funds to purchase a new automation system or upgrade when the state is offering one for free. Quality has little to do with political reality. Is Evergreen a good thing happening in the profession or is it just something we want to believe is good?

You can really speed up your circulation by putting the barcode on the front of the book. Your current method is inefficient. I've clocked many a circulation clerk who can checkout 10 books to a patron in under 10 seconds on a consistent basis.

Is Evergreen a good thing happening in the profession or is it just something we want to believe is good?

What do you mean by this?

The creation of a viable (not perfect maybe), but viable alternative to closed, expensive, vertical options where getting at your own data is difficult, or at least expensive is a good thing for this profession. How could it not be?

Evergreen could be good if:

it were not simply an attempt to replicate existing systems.
it significantly bettered the existing systems.
it was actually a cost savings.

Mediocrity won't shake the market or cause any noticeable effect.

it were not simply an attempt to replicate existing systems.

You have either not looked at Evergreen's infrastructure and design or you're ignoring it. The OpenSRF framework is head and shoulders above any other ILS offering's underpinnings. Based off a solid foundation, one can do anything. I'm not going to enumerate the various benefits or features, you can research it yourself.

it significantly bettered the existing systems.

Similar response. Add in the benefit of open code, an open and robust API (a real API, unlike what many ILS's have), the ability to run on inexpensive hardware in a clustered environment for fail-over and redundancy, the ability to simply add more hardware for more capacity (on the fly), upgrades to the server-side (on the fly, no downtime), the ability to properly model organization hierarchy, a granular permission system based on individual login, location, and proximity... Gosh, I did some of your research for you.

it was actually a cost savings

Actually, in that spreadsheet, some of the numbers in the column on the left are for half of the fiscal year (since many of the services were canceled in Dec '06. So, yes, the numbers are a little "suspect")

Let's look at it with the real annual numbers, shall we?:

-PINES net gained 2 staff for software dev and support. (+$200k per year)

-PINES had to buy a new hardware (+$300k one time cost)

-PINES canceled Sirsi support (-$200k per year)

-PINES canceled Sun support (-$100k per year; support is included in above new hardware cost for 5 years)

-PINES reduced server housing costs (-$100k per year)

-PINES did not have to purchase a new server for their old ILS, the old server was purchased in 1999, and was at EOL (-$1,500,000.00 one time cost, if I recall correctly)

Where's your leg to stand on?

Load balancing mechanisms are hardly new. More efficient to accomplish in hardware. It still checks books in and out to patrons, so no matter what the infrastructrure is it ain't exciting nor innovative. Innovation needs to be front of the house.

Well it's open source, but it isn't meeting the needs. Just adding cost. Everyone wants the David vs Goliath tail to be true. Just don't hold water.

As for the numbers: The biggest problem is that Evergreen development was charged to the PINES budget while supporting Sirsi. So the cost savings only are the reduction of Sirsi and associated costs. Work your numbers backwards and correct FY07. When thats done FY07 comes in under a million. Seems all the cost savings went away when Equinox became a company. FY08 with network costs and staff added in will top 2 million. No economy of scale.

From the 2007 Access> in beautiful Victoria, British Columbia, we wish to thank you for the comedy fodder your postings have provided us. From utter ignorance of Evergreen (somehow comparing a hardware load balancer to OpenSRF...), to weaving a tall tail, to being incapable of 2nd grade math... our hats off to you dear sir (or madam). /1900015790.html?>

How is it clear that Evergreen is not the culprit? Just because the Evergreen folks says so? Because you say so?

I would think that the first thing that anyone would look at would be the connectivity issues, and I include both the Evergreen folks and the libraries who are complaining. You are insulting the intelligence of these libraries when you "assume" they didn't bother to check the connectivity. It's possible that the majority of them may not be as techno-savvy as the rest of the world, but it's rather rude to assume that all of them are idiots (as you imply) when it comes to technology. It's amazing how many libraries seem to have "connectivity issues" that didn't have these issues while on the previous ILS vendor with the same equipment. Of if they did, they weren't as vocal about it.

And, why would anyone have a grudge against Evergreen? Please. What I have issue with is someone not addressing these problems and just tooting the horn of "open ILS" as the end all be all to PINES and other library system's ILS problems. The slowness claims are legion throughout the PINES system, and complaints have been made from week 1 of the PINES implementation Sure, you can't please all the people all the time, but if a significant portion of my state library money is going to support this automation system, then I want one that works and works well. There are much larger consortiums than PINES that are running just fine on commercial ILS systems and don't seem to have the connectivity issues that PINES libraries experience.

Also, why do you hail open-source as the second coming to the profession? Particularly this open-source vendor? What gives you any idea that these people know anything about working a circulation desk? According to their bios on their website, none of them have spent any time actually working in a library. Yeah, yeah, they worked for the State Library, but have they ever been standing at a circulation desk, in the middle of Summer Reading, looking at a sea of patrons, 40 deep, all wanting to check out 15 books at a time? And, you can't check out the next set of books, until Evergreen decides to let you finish? If I am incorrect about their experience level, I do apologize, but according to their posted experience, they aren't all that qualified to make the kinds of judgements about what is slow and what is not at the circulation desk.

Just my anonymous .02 worth.

No matter how you slice it Evergreen costs a third more than the previous system.

I would love to know why the cost of supplies jumped 4,100% from FY07 to FY08.

Some new figures from a PINES presentation />

Check out slide 11. For those who don't want to download I'll paste the text and remove whitespace
Cost effectiveness:
Estimate to replace automation systems in all
PINES libraries: $15 million
Annual maintenance for
Individual automation systems $ 5 million
PINES annual operations: $1.6 million
Approximately $1.00 per registered customer
About 1/10 of the cost of purchasing individual automation systems

15 Million to replace the existing system? Seems a little high I know that figure comes from at least 3 years ago but come on does anyone really believe it would cost $312,500 to convert each system? Those are small libraries.

$5 Million in annual maintenance? $104,166 for maintenance for each system? Thats a third of the purchase price on maintenance. Get real.

1.6 Million for Evergreen. Well thats a new figure with the original system costing 1 millon to maintain that means open source Evergreen is a 60% more expensive? Oh thats a savings.

$1 for each registered patron. Thats a deal? If it's more than a dime for the licensing for each registered patron it's a bad deal.

Actually 1.6 Million would be a good support deal if you spent 15 Million. Personally I have a hard time believing that you couldn't do this entire project for 5 million. Which means paying 1.6 Million would be about a third for maintenance.

Maybe some money spent on someone who understands negotiation would have a pay off. Where can I send the invoice?

One last note. These figures still do not include GPLS personnel expenses. I have a suspicion that the 15 million figure also included rolling out state wide internet access, buying all new computers, new periphereals, along with salary for all libraries. Problem with including that is all of that is reusable.

Shame shame as they say "figures never lie, but liars always figure". How are sane people sitting in these presentations without hearing a cacaphony of bullshit coughs?

I really think a price tag of 3 Million would have been reasonable but certainly no more than 4 Million. Whatcha wanna bet that the price is going to keep going up? Meet the new vendor same as the old vendor.


Couldn't have said it better, Anonymous Patron!

Well, that's really funny, because I AM a librarian from Georgia and I work in a library that uses Evergreen and our circ desk certainly checks out hundreds of books an hour... certainly more than 20 an hour! I think the Pines system overall performed more than 19 million circs last year!

I can definitively say that we love Evergreen and it works great for us. The folks who wrote it really listened to our needs and designed a system that would work in tiny rural libraries with a few staff, medium-sized libraries (like mine), and it also works well in larger libraries. The software is so flexible, powerful, and intuitive. I don't even think about "if" it will work when I walk into the library in the morning because it just does. That can't be said for our former system and I know there is no software on the market (proprietary or open source) that can do what Evergreen does, day in and day out.

I can't speak as to why this particular library did not chose Evergreen. I'm not involved in that level, I'm just a lowly librarian in the field, I don't even know if they contacted Equinox Software or the State Library of Georgia for information. It could be the Evergreen folks don't market very well, and frankly, the guys who wrote the software are some of the most humble geeks you'll ever meet. They don't toot their own horns enough. But boy does LibLime ever toot their horn for every little customer they get!

I think the offensive nature of the first anonymous coward's comment was actually directed towards Koha - the use of "it's", as an indefinite pronoun, might have been intended to refer back to Koha rather than the nearest noun Evergreen. Trolls typically aren't known for their grammatical skills.

No matter. Either way, the first anonymous coward's comments appear to be directed towards the purpose of sowing dissent between the two most visible open source library systems, which (if the intention succeeds) would simply waste energy that could be put towards improving each respective system. Let each system compete on their technical merits, and let them collaborate where possible, and we all benefit. Throw FUD, slick marketing, and outright dismissals of the merits of either system into the mix, and we all lose.

I'm kicking myself, because I just lost ten minutes I could have put towards a more productive purpose - but I do hope that this helps direct any negative energies the comment stirred up towards something more fruitful and real.

A conspiracy theory predicated on grammar? For a second I though I was reading the Onion. Personally my read on the comment was this is a big compliment for Koha and LibLime. There is a poke at Evergreen, but the question is valid. Why choose the "little library system" for a library with tremendous circulation instead of the "big library system". Though 19 million in circulation does raise my curiosity. If true one library in Maryland is circulating 25% of the entire state of Georgia! My last thought was wondering how someone could spend 10 minutes on such a silly post.

Though 19 million in circulation does raise my curiosity. If true one library in Maryland is circulating 25% of the entire state of Georgia!

Pines does not include several large GA libraries at this time:

Atlanta Fulton: ~3 million circ
Cobb County: ~4 million
Forsyth County: 1.5 million
Gwinnett County: ~7 million

If I read that correctly 4 libraries almost match the circulation of the rest of Georgia? Apparently PINES has 270 libraries that can barely out do 4. When you see the publicity on PINES it leads you to believe it's the entire state of Georgia.

Seems pretty clear to me:>

Everybody working in a PINES member library knows that Evergreen is SLOW. Too slow for libraries like Atlanta-Fulton and Gwinnett.

Heresy. Evergreen is the best system in the world... if you have all day to wait in the check-out line.

Mr. Scott, shouldn't you acknowledge that you are a contributor to the Evergreen effort? That you have a vested interest in the success of Evergreen?

Mr. Anonymous Patron:

Vested interest? I created an account for the explicit purpose of posting a comment under my real name (which is pretty heavily associated with various open source projects, including Evergreen) to avoid accusations like that. Ironic.

I will acknowledge that I could have misinterpreted the original comment's intentions, and that they might indeed have meant that Evergreen was slow. To date, the bulk of my experiences with Evergreen have been running the entire stack on a 512 MB virtual machine. Given its performance under those situations, I find it hard to imagine that scaling that configuration up to a clustered database server / web server / messaging server would be slow. But okay, let's explore the possibilities, find out if there is truth in the assertion; if so, find out where the choke point is, and figure out how to fix it.

One of the nice things about an open project is that the bug database for Evergreen is open. Running a search for "slow" against all bugs in any state returns four results. The most on target report seems to associate the slowness with the speed of the library's connection to the net - however, there is no resolution to the bug so it's not clear if that was the real reason behind that bug report.

One of the other bugs has more to do with the number of mouse clicks required to perform a given task - a usability issue, to be sure, but one that could be addressed if the use case warrants it.

The other two bug reports didn't seem to be closely associated with the slowness of the system - but I invite you to go and look at them yourself.

Part of the problem with casting the aspersion "slow" against a system is that it's not clear what speed is being measured, and what all of the factors are. Is the client running on a PII with 256 MB of RAM? Is the connection a dial-up connection? In the database world, there are a set of standard benchmarks ( that measure various database use cases, and require published benchmarks to be audited and scrupulously documented, so that one can actually measure what is slow.

So two things:

1. It would be great, if there are Evergreen libraries still experiencing slowness today, if those libraries could open bug reports so that the causes could be determined and the resolution of the situation recorded.

2. It would be wonderful if some organization would be willing to put together a library systems parallel of the TPC benchmarks so that there could be empirical measurements of claims of performance for specific scenarios.

Why is it in a profession that strives to preserve anonymity so many practitioners have a problem with respecting the desire to remain anonymous? Many reasons why people choose anonymity, most are not nefarious. The fact of the matter is everyone is anonymous on this discussion. There is no verification of identity upon registering. When you login you supply the unverified credentials. If you want to paint yourself as self righteous because you have a login under your name have at it. This anonymous patron isn't impressed.

Just what is the relationship between the state of Georgia and Equinox? I've done a little digging and Equinox's management team is made up of the same people who developed Evergreen (former state employees). Quite the incestuous relationship.

Most of the time you see a Koha announcement and it turns out that the library doesn't need automated circulation. This one is different, NES has the 2005 circ at 4.7 million that qualifies as real. Curious why they didn't choose Evergreen. Isn't it the big library system? Friends in GA on Evergreen tell me it's the perfect system if you plan on checking out less than 20 books an hour. Hats off to Liblime for getting a real customer.

You didn't have to dig very hard. The fact that we at Equinox are the original developers of Evergreen is on the front page of our website at> and it's something we discuss at every presentation we've given since July 1 -- the date we were hired by the Board of Regents of the State of Georgia to provide support and development services for Evergreen.

As for anonymity, it speaks to the veracity of claims made that one cannot back their statements with information that allows those statements to be followed up and verified. I invite all manner of lively and mature discussion with those willing to stand by their claims, but I tire quickly of those who find it intellectually acceptable to snipe with FUD from bushes. If any of you out there would like to discuss any of the unsubstantiated claims further, feel free to contact me using your real name directly at miker[at]esilibrary[dot]com, stop by our IRC channel on freenode at #openils-evergreen, or join the Evergreen mailing lists available at> .

Congrats to LibLime on their contract. Success for Open Source in libraries is a success for the profession and for library patrons everywhere.

-- Mike Rylander of Equinox Software, Inc.

I have been silently watching the comments posted on this story up till this point, but lately they are simply wrong and mean spirited and I have to say something.

I'm a staff member in a metro Atlanta (non-PINES) library but I live in a neighboring county served by PINES. My daughter and I frequent the local PINES library, (Piedmont Regional Library System) but I have never seen problems or the purported slowness.

I don't work in a PINES library so I may not be as intimate with the issues as some, but in my experience, everything has been fine. I have spoken with various staff members in my local library and sure they have concerns and problems, but overall they like Evergreen.

So I cannot figure out who these anonymous people are posting nasty things about Evergreen and Mr. Scott, who is just trying to help with useful questions and comments. Are you all employed by some competing ILS company trying to smear Evergreen out of fear?

Or if you are really having a problem is this the correct forum to air those concerns? It just smacks of unprofessionalism. Say, does your director know you are posting this garbage? If your check-out lines were really that backed up, should you not be working the desk instead of writing nasty comments here?

Maybe you should pay a visit to Piedmont Regional in order to learn how to run the system properly to better serve your patrons? Or, as Mr. Scott suggests, investigate your network to determine if thats causing the speed issues and FIX IT.

I believe PINES has expanded by at least a couple library systems recently. So where are the libraries leaving PINES because the system is so slow and bad? The facts don't add up. I know the system at which I work is strongly considering Evergreen, but maybe not PINES, after they develop the Acquisitions module.

This is very childish, and I agree with Mr. Scott in that I cannot believe I just spent 10 minutes responding to this utter rubbish. You folks need to get a life. I also agree with Mr. Rylander, congratulations to Liblime.

If I read that correctly 4 libraries almost match the circulation of the rest of Georgia? Apparently PINES has 270 libraries that can barely out do 4. When you see the publicity on PINES it leads you to believe it's the entire state of Georgia.

Not sure what your point is. 19 million in circ is big, whether it is the whole state or not. Queens Borough does 21 million, and is considered the busiest system in the nation. The point is, it is impressive enough that there is a system that can handle the complex needs of so many individual entities, while still handling a transaction load close to that of the busiest library in the country!

Mano y mano... via email. What a hoot! For speed issues I suggest looking at the pines circulation list.

1) Piedmont has pretty low circulation. I doubt the speed issue is a problem there. The speed issue is a problem for a busy library. If your not busy it doesn't matter.

2) Correct forum? They aren't listening to their own forums so if this gets a problem fixed I'll use it. Does my director know? Who says I'm not the director! Would your director be happy with you posting during work hours?

3) Fix it? If I could fix it I would. Yeah I like being slow with cranky unsatisfied patrons. The speed issue has been well documented. The network is the latest scape goat. First it was the computers were to slow, then not enough ram, then your local network is broken, then its a priority problem, then you need more bandwidth, and finally we need more bandwidth to the server. Changes to all have been made. Still slow, no improvement. I can circulate an item at 8am and still see the speed issue. Workflows was many things but in comparison greased lightning. Another thing Workflows seemed to handle very well was receipt printing much much faster than Evergreen.

I think the point is that PINES isn't as big as we were led to believe.

You just said two opposing things: a) it's only an issue in a "busy" library b) it's slow at 8 in the morning. Which is it? Unless the system magically knows the checkout is coming from one of those special "busy" libraries and decides to slow down just for you...

I'm not an IT person by any stretch, but it sounds like someone is trying to help, but they're not sure what the issue is. In any case, do you think they'd spend money expanding bandwidth and the other things if they knew it was the software? I would think it would be cheaper and faster to fix the software or get a faster server than run a more telecom lines for you?

What I'm trying to say is that it seems obvious to everyone except you that it's not Evergreen itself. If I can stand in a Piedmont Regional Library, and the checkout goes as fast as the staff member can scan the books, and you're standing in your library and checkouts are dragging... come on, connect the dots here. Unless I fundamentally misunderstand how PINES is set up... there is one big central server you all connect to, right? So, the only difference would be your library's network, the computer you're checking out on, and the network between you and the PINES server. That's the logical conclusion.

I was at a presentation a couple months ago and the PINES budget was given including what Equinox is being paid. This is not the exact presentation, but it has the slide I recall. EVERGREEN/SCKLSPINESEvergreen06-2007Presentation.p>

Go to slide 16. According to that, technology costs have decreased and other costs have increased, especially this courier service line. I assume that's for transporting books to fill holds. Equinox's line is $350,000. That looks like it pays 4 people who also have to pay their own benefits and insurance since they're contractors. I wonder how much people that can write an ILS make normally? Gravy train? Where's the beef? What do you have against these guys?

Not opposing: Piedmont has low circulation... ie fewer people standing in line to checkout an item. Hence a 2 second wait isn't a problem. Add up all the additional 2 seconds per transaction then add in a busy library. They can only add X amount of people to the circulation desk to compensate for the slow down caused by Evergreen. Two seconds for the patron scan, two seconds per item followed by a 30 second wait to print the receipt.

That slide is missing a few key bits of information. Additional staff at GPLS. Yes the 4 programmers are now contracted as Equinox, but 2 of the vacated positions were filled. Network capacity. Many libraries have received additional T1 circuits for Evergreen speed and now the entire network infrastructure is being upgraded to support the additional network needs of Evergreen. So FY2007 should be much higher. Figures for FY2006 would be much smaller if the following to support Evergreen development were subtracted: 2 staff members, server clustering, CCG contracting. Those facts alone would drop the price to around a million dollars. This math on that slide is suspect. Probably should read FY2006 1Million, FY2007 2Million.