Why would you charge for Datasets/Crossgrade and additionally for subscription?Because the new datasets are often 'never complete' and need ongoing maintenance and expansion.
(Mark, this would be Category 2.5 in your list -- a dataset that gets continual updates from the server.)
The Logos Cultural Ontology Dataset, for example, initially came with tags for the Bible and a few other ancient texts. We've been expanding the tagging to other texts continually since then. It costs just as much to put another year of tagging into it as it did to put the first year into it.
The Bible was obviously at the top of the list, but the whole dataset gets incredibly more valuable as more and more obscure texts are added -- you can then use it to find cultural parallels and references that you might never have thought of. But while it's easy to say this dataset for the Bible is worth $19.99, (I believe that) it's hard to get people to make another explicit transactional purchase to add the tagging for just the Nag Hammadi Library; but as we add Nag Hammadi, Ancient Egyptian Literature, Legends of the Jews, etc. the dataset becomes much more valuable. (And it costs as much, if not more, to tag these texts as the original set.)
And it's never done -- we plan to extend the tagging to the Talmud, the Mishnah, the church fathers, Pliny, the Iliad, etc.
And this is just one dataset -- we're doing this kind of extension to many. It's hard to describe and offer each extension as something new to purchase, and logistically difficult to charge for these things. (Will some users have Iliad tagging and others not? How will tech support help people who don't realize they've only purchased four sevenths of a dataset?) We might end up holding a years or two worth of data to make the value perception high enough, which delays delivery to everyone, and still is complicated.
Sell me Logos 7 today!
Comments
-
Bob Pritchett said:
t's hard to describe and offer each extension as something new to purchase, and logistically difficult to charge for these things. (Will some users have Iliad tagging and others not? How will tech support help people who don't realize they've only purchased four sevenths of a dataset?)
Hmm - the question that rises in my mind is WHY is it so difficult.
For 'new purchases' it's just a matter of valuing/ pricing surely.
For example (again wildly exaggerating to make the point that it is ONLY pricing that is the issue!) if a customer paid $25,000 for lifetime un-removeable access to this dataset, then what is so difficult? It would extremely easy - no 4/7ths, no 'logistically difficult'-ies to be able to charge! In fact if someone offered that it would probably be SO easy it guess it would be a 'no-brainer'!
Now $25,000 may be an exaggeration (the again - you might believe that it IS worth that much!), but the point is that it's just a matter of agreeing a price (i.e. value for the dataset/feature over a lifetime). You need to determine the 'lifetime value' of each feature / dataset for owners.
The account page of the customer will have "paid: Yes' and that is that!
PS: As for tech support, again not difficult - you should charge for that as per your previous suggestion and I would assume the 'customer account screen' that your tech support access will have all relevant details of what is owned / rented etc (for how long etc)
0 -
James Hudson said:
You need to determine the 'lifetime value' of each feature / dataset for owners.
Having lived long enough to have seen several economic cycles, your simply is my complex ... the value of a dollar (or other monetary unit) varies over time. That's why you need actuaries to calculate annuity rates. And Bob's costs will go with the fluctuating dollar while the customer who bought a lifetime rental (sales are not limited to lifetimes) get to pay an estimated present value.
Orthodox Bishop Alfeyev: "To be a theologian means to have experience of a personal encounter with God through prayer and worship."; Orthodox proverb: "We know where the Church is, we do not know where it is not."
0 -
James Hudson said:
Hmm - the question that rises in my mind is WHY is it so difficult.
For 'new purchases' it's just a matter of valuing/ pricing surely.
One problem is that at the beginning of the project you don't always know how popular/big the project will get. You might start with adding tagging to a particular corpus (the Psalms, for example), and it ends up being very popular so you extend it the the Wisdom literature. Customers then find new ways of using the material so it extends again to the whole OT, and then again to the NT, and then again to the deutrocanonical texts, and then again to the LXX, and then… well, you get the idea. Of course, you can easily create separate datasets for everytime you expand it, and sell them as extras, but as Bob says, that is complex. Logos Now is simple.
Conversely, under the Logos Now system, Logos can pull the plug on datasets that don't prove popular or useful. The roll it out for the Psalms, but no-one uses it, so they don't expand it. Under a strict 'sales' approach, they would have to continue developing the set, even if only a handful of people had actually bought it (or issue refunds, etc.).
In other words the subscription method gives Faithlife a great deal of flexibility. They're only promising we'll get good things in the future, they retain the freedom to choose what they are. The purchase model ties their hands. Sometimes that's OK. Sometimes it stifles innovation and prevents 'risky' but potentially valuable projects. That's one reason why Logos uses both models.
This is my personal Faithlife account. On 1 March 2022, I started working for Faithlife, and have a new 'official' user account. Posts on this account shouldn't be taken as official Faithlife views!
0 -
Mark Barnes said:
One problem is that at the beginning of the project you don't always know how popular/big the project will get.
If you (Logos) build it, they will come.
george
gfsomselיְמֵי־שְׁנוֹתֵינוּ בָהֶם שִׁבְעִים שָׁנָה וְאִם בִּגְבוּרֹת שְׁמוֹנִים שָׁנָה וְרָהְבָּם עָמָל וָאָוֶן
0 -
Mark Barnes said:
under the Logos Now system, Logos can pull the plug on datasets that don't prove popular or useful
I might have missed something, but this doesn't sound like either Logos Now or what Bob would do...
Firstly, I don't know how Bob would KNOW which of the LN items were popular and useful - as I understand it, LN is a package which Logos have committed to adding new features/sets every 6 weeks - so this is effectively one package. How would they know which elements were proving most popular?
Secondly, I've read nowhere about them REMOVING elements from the package (I might have missed the announcement), but that could make the $9pm payers upset. It doesn't necessarily sound like something Bob would do - he always tries to do well by the customers.
If this is true, do all the people who currently pay their $9 pm know that Logos may be removing features.
Mark Barnes said:Sometimes it stifles innovation
Not entirely sure how? Their commitment to a new feature every 6 weeks (approx 8 pa), which means after 10 years 80 new features etc seems to be a drive to innovation to me!
Mark Barnes said:Logos uses both models
The point is, they don't use both models for LN!
Thanks for you comments as always Mark.
0 -
James Hudson said:
How would they know which elements were proving most popular?
Usage statistics - at the open house we got a quick peek at the statistics IIRC sentence diagrammer was near the bootom - I would suspect because of its usability issues rather than user desire ... but statistics don't give the why.
James Hudson said:Their commitment to a new feature every 6 weeks (approx 8 pa), which means after 10 years 80 new features etc seems to be a drive to innovation to me!
Where did they say that? The updates may be broaden dataset not new features ... or redesigned, improved features as several releases already have had as a strong focus ... or a step-wise implementation of a complex feature as we saw with concordance which I would claim is still unfinished because it doesn't support phrases (n-grams).
James Hudson said:The point is, they don't use both models for LN!
Logos Now is specifically a combined owned/subscription model. Logos Cloud is subscription only model and Logos is ownership only model ... with porous boundaries i.e. actual users fall on a continuum.
Orthodox Bishop Alfeyev: "To be a theologian means to have experience of a personal encounter with God through prayer and worship."; Orthodox proverb: "We know where the Church is, we do not know where it is not."
0 -
James Hudson said:
I might have missed something, but this doesn't sound like either Logos Now or what Bob would do...
(...) I've read nowhere about them REMOVING elements from the package (I might have missed the announcement), but that could make the $9pm payers upset. It doesn't necessarily sound like something Bob would do - he always tries to do well by the customers.
If this is true, do all the people who currently pay their $9 pm know that Logos may be removing features.
I'm not sure if you are misunderstanding Mark here. Especially with the example given by him, I understood "pull the plug" to mean 'stop further development', but not 'remove existing feature' and I'm sure Bob does that all the time - if not explicitly then implicitly by approving priorities which will in effect lead to the A prios to be done and the C prios to remain untouched.
Have joy in the Lord!
0 -
NB.Mick said:
I'm not sure if you are misunderstanding Mark here. Especially with the example given by him, I understood "pull the plug" to mean 'stop further development', but not 'remove existing feature'
That's exactly what I meant.
(Although I'm sure features will be effectively removed from Logos Now in the future — because they'll be made freely available to everyone.)
This is my personal Faithlife account. On 1 March 2022, I started working for Faithlife, and have a new 'official' user account. Posts on this account shouldn't be taken as official Faithlife views!
0 -
James Hudson said:
Not entirely sure how? Their commitment to a new feature every 6 weeks (approx 8 pa), which means after 10 years 80 new features etc seems to be a drive to innovation to me!
They don't sell a new feature every six weeks. That's not a purchase model.
With the Logos Now model, Logos has to include enough valuable content in the package to ensure people sign up to it. But once they're signed up, they can afford to experiment with at least some of the new content. A project can be given a budget of (say) 10% of Logos Now subscription money for the next three months, and produce their innovation with that budget. If users like it, maybe the project gets extended. If they don't like it, maybe it stays where it is. But at least it gets funded. You can take the risk of people not liking it.
On the other hand, the purchase model can stifle innovation because with that model, Logos only develops things it knows it can sell (the Pre-Pub and Community Pricing schemes are the best example of this). If no-one buys it, it doesn't get developed at all. That's often a great model, but it occasionally fails because sometimes we don't know we want things until we have them. Logos Now is great for those scenarios.
This is my personal Faithlife account. On 1 March 2022, I started working for Faithlife, and have a new 'official' user account. Posts on this account shouldn't be taken as official Faithlife views!
0 -
Mark Barnes said:
Logos only develops things it knows it can sell (the Pre-Pub and Community Pricing schemes are the best example of this). If no-one buys it, it doesn't get developed at all. That's often a great model, but it occasionally fails because sometimes we don't know we want things until we have them. Logos Now is great for those scenarios.
Mark,
Is there a list of "things" or projects known in advance to Logos developers? Are they developing things by air, hoping the new features would sell? Do they know where they are going before they get there? Your statement: "If no-one buys it, it doesn't get developed at all... it occasionally fails because sometimes we don't know we want things until we have them", appears to be a "hit-n-miss" plan of developing new features. Who's driving the train, here? Is it all Faithlife or is it all Logos users? Can it be a combination of the two? If I am wrong, I stand to be corrected.
Is it possible for Faithlife to reveal the list of future projects and have Logos users give (general thumps-up) a sense of interests and what features and/or books would fly? This way, it would same time, money, energy and give a sense of certainty of what's marketable. It's scratching people where they itch and ensuring Faithlife's bottomline. Looking down board is not a bad thing.
IMHOP, if Faithlife comes up a future list of developments or ask users of their interest, the same users can help prioritize list to avoid the "hit-n-miss" highway of doing business . This would same time, make them money, and be assured of happy customers.
0 -
Mark Barnes said:
I may well have misunderstood - I apologise.
But if it is the case that Logos may "stop further development", that that is not a good argument against having a purchase model for LN datasets/features. If the feature/dataset is not worked on any further, and not removed then it could have still be purchased. (There would be no need for refunds etc just as in the same way that under the rental model Logos would not need to reduce the $9pm, and there would be no need for refunds etc).
If nothing is going to be removed, then it is in effect JUST a way of paying - either up front or monthly.
I'm sorry if I am misunderstanding something, but it seems clear to me so I am obviously not explaining myself well enough. Thanks for bearing with me. I'm just wanting to somehow give Logos my money, but not under a rental model!!
0 -
MJ. Smith said:
I'm sure I read that somewhere. But can't find out where! So I've probably got that wrong. Having worked with software developers, I did think it was a bit ambitious!!
[:D]
0 -
Charles McNeil said:
Your statement: "If no-one buys it, it doesn't get developed at all... it occasionally fails because sometimes we don't know we want things until we have them", appears to be a "hit-n-miss" plan of developing new features. Who's driving the train, here? Is it all Faithlife or is it all Logos users? Can it be a combination of the two?
From your questions, I suspect you are not familiar with the technological state of natural language processing and text analysis (unstructured data mining).There are several thread that feed into what gets developed in Logos:
- items needed to support current teaching trends in Bible colleges and seminaries.
- items requested by users
- items that extend current functions - especially items that were not but are now technologically feasible
- items needed to remain a leader in the field i.e. customer expectations
- items that the scholars-in-residence at Logos have on their dream list that have finally become technologically possible or they have found a way to show its usefulness to their users
I've not seen their development plan but I would be sure that it is a standard IT plan ... one that get revised frequently as issues arise and priorities change. Remember that academic research is a trickle down system - the originators spread it to a few who start spreading it among graduate students who start spreading it among undergrads who spread it to the overall population. And most students stick with the methodologies that were popular when they were in the classroom. Only a few keep updating their skills.
Orthodox Bishop Alfeyev: "To be a theologian means to have experience of a personal encounter with God through prayer and worship."; Orthodox proverb: "We know where the Church is, we do not know where it is not."
0 -
Charles McNeil said:
Your statement: "If no-one buys it, it doesn't get developed at all... it occasionally fails because sometimes we don't know we want things until we have them", appears to be a "hit-n-miss" plan of developing new features. Who's driving the train, here? Is it all Faithlife or is it all Logos users? Can it be a combination of the two?
I am part of the team that manages the list of features for Logos on the desktop, so I can vouch for the existence of the list, or rather, lists.
There's an ideas list, which has suggestions from every imaginable source from hare-brained to half-baked to genius. The items on the various lists are essentially as MJ suggests.
Then there's the production backlog, which has all the things we're actually planning to do in the order we're planning to do them. I tag them with the following categories, ranked from least to most interesting:
- Bugs, which have their own triage system based on severity, spread, and fixability.
- Plumbing items necessary for the system to continue to function such as keeping up with new OS updates, upgrading development tools, or maintaining servers.
- Internal items necessary to support dataset and resource production such as the tools we build to create reverse interlinears, and various compilers for all the different types of resources and datasets we build.
- Upkeep items necessary for fixing bugs and design flaws in existing features for greater utility and usability.
- Improvement items that extend the functionality of current features beyond their original charter.
- Refit items that are completely rebuilding an existing feature from the ground up, such as the recent refitting of the Information tool.
- New features and datasets that add previously unknown capabilities and content to the platform.
Logos Now only consists of items from the last category, because we don't charge a subscription fee for bug fixes or maintenance.
The production backlog is ranked using a conglomeration of marketability, perceived value, estimated utility/usefulness, absolute necessity, etc without much thought to feasibility.This process consists mainly of a series of inferences and extrapolations (read: educated guesses) from whatever limited information we have.
Nowadays we plan out each 6-week cycle release by building a list taken from the production backlog, favoring the highest-ranked items that are also feasible, that is, that we can execute with available technology and people-power.
We try to balance all the categories above. We can't decide not to do some plumbing tasks, or the whole system collapses. We can't decide to just focus on new feature development and leave upkeep and improvement of existing features by the wayside, or longtime users will be unhappy.
We revise the production schedule as more information becomes available, sometimes daily.
0 - Bugs, which have their own triage system based on severity, spread, and fixability.
-
Charles McNeil said:
appears to be a "hit-n-miss" plan of developing new features. Who's driving the train, here? Is it all Faithlife or is it all Logos users? Can it be a combination of the two?
It's not a "hit-n-miss" plan so much as the way free markets have operated for thousands of years. [:)]
And it's definitely a combination of the two.
Development effort is costly and so we are highly incentivized to pick only winners and sort out the losers when we make our production schedule. But the only sure way to sort the winners from the losers is to bring our best guess at the list of winners to market and let consumers vote with their feet. Ultimately, that's the only measure of value that matters.
Charles McNeil said:Is it possible for Faithlife to reveal the list of future projects and have Logos users give (general thumps-up) a sense of interests and what features and/or books would fly? This way, it would same time, money, energy and give a sense of certainty of what's marketable. It's scratching people where they itch and ensuring Faithlife's bottomline. Looking down board is not a bad thing.
IMHOP, if Faithlife comes up a future list of developments or ask users of their interest, the same users can help prioritize list to avoid the "hit-n-miss" highway of doing business . This would same time, make them money, and be assured of happy customers.
We have considered a sort of pre-pub system for features in the past, but it hasn't passed muster for a variety of reasons: New book development can be estimated to within a standard deviation or two, but new feature development cannot. The value proposition for a new book is simple for us to propose and simple for customers to evaluate as a purchase decision. This isn't as straightforward for software features — less so the more innovative the feature is. Except for a few outliers, new books are equally feasible in terms of the technology and people-power they require. Software features are not.
In the meantime, every feature doesn't stand on its own merits but is bundled with all the other features, hopefully in such a way that the whole is greater than the sum of its parts. The value of the interconnection among features is an emergent property of the program as a whole, not of an individual feature. Indeed, many features have no value except when interconnected with other features. More importantly, the 10% of the features you use is subsidized by the 10% of the features someone else uses.
(Voting on features individually is a bit like the inititiave process for making laws. In Washington state, which has a robust referendum and initiative process, we the voters often vote FOR services and AGAINST the taxes which pay for those services. Then we leave the government with a nice mess to sort out instead of the other way around, which I guess is a refreshing change of pace!)
Because of that bundling, we can't just use purchase information to know what parts of the software are outperforming others. So we also use the telemetry (controlled by the "Send Feedback" program setting) to help us answer: Once they've bought it, do they actually use it? We also factor in reviews, blog posts, forum posts, social media interactions, customer service reports, etc.
If no one wants to use the thing they've bought, then good stewardship tells us we need to put our resources elsewhere in order to focus on what brings the most value to the most people. (That is, unless we know for a fact that the thing isn't getting used because it needs further design and development to have good utility for more people, in which case we are proceed with caution so as not to throw good money after bad.)
Otherwise, we'd go out of business quickly, and rightly so.
So, bottom line: It's Faithlife who determines what to build, based on our best guesses of what users will get great value out of (see my previous post). We are in that sense "driving the train." But it's the aggregate of all Logos users who decide how much value they assign to the bundle of features (price, purchase decision) and how much they actually utilize each individual feature (usage statistics). So you get to decide whether or not you want to ride the train to where we're driving, which we in turn factor into the decision of where we decide to drive next.
This creates a sort of dialectic or spiral or feedback loop (whatever you want to call it) that, as I say, drives markets forward.
From my perspective as a project manager, subscriptions coupled with shorter release cycles are helpful because they bring the purchase decision and the production decisions closer together in time, which makes it much easier for us to respond to customer feedback and pivot our plans as necessary. It tightens the loop, as it were.
None of that should be controversial. Apologies for being dense, but I've lost track of how any of this applies to the original discussion. Then again, please don't feel obligated to bring me up to speed, either. [:)][:)][:)]
Happy Thanksgiving!
0 -
All I want for Christmas is for this thread to end. We have beaten a herd of horses until they were dead twice[:O][;)]
0 -
Fredc said:
All I want for Christmas is for this thread to end.
You and me both. (You'd think I'd stop reading it though ... [*-)][:$])
0 -
macOS, iOS & iPadOS |Logs| Install
Choose Truth Over Tribe | Become a Joyful Outsider!0 -
As the original poster I'm satisfied that Faithlife has listened and considered my request. Seems they don't agree as is their choice. As a customer I value that they have listened. I'm staying along for the ride.
I didn't intend to start such a long thread but I guess it struck a chord... I'm still grateful for the software and library, which I find very helpful. Happy Thanksgiving all!
גַּם־חֹשֶׁךְ֮ לֹֽא־יַחְשִׁ֪יךְ מִ֫מֶּ֥ךָ וְ֭לַיְלָה כַּיּ֣וֹם יָאִ֑יר כַּ֝חֲשֵׁיכָ֗ה כָּאוֹרָֽה
0 -
0
-
John Goodman said:
As the original poster I'm satisfied that Faithlife has listened and considered my request... I didn't intend to start such a long thread but I guess it struck a chord... I'm still grateful for the software and library, which I find very helpful. Happy Thanksgiving all!
John,
No apologies needed. You raise the right topic, in the right forum and in the right thread. Those who don't care for this thread can do one or four things:
- Don't tune in.
- Read it and don't make any comments.
- Take interest in other threads
- Read the many books in their library.
The answers are obvious, but the questions must be asked:
- Are Logos Users required to use this forum?
- Are they restricted to whom and what thread they can respond to?
- Is there a limited frequency of usage?
- Is one required to respond to a statement in a thread he or she doesn't agree with?
You're an Ok Brother. Have a Happy Thanksgiving, wherever you are!
0
