• Welcome to AppraisersForum.com, the premier online  community for the discussion of real estate appraisal. Register a free account to be able to post and unlock additional forums and features.

Dump Adobe, Use PDFCraft

I am sure that Adobe will take legal action and do what they can do to cause problems for ArtCraft ....

I would advise everyone to download the latest ArtCraft every day. It is very good - thanks to Rust. But hard to say if the rug gets pulled out from under them.

I already have (in less than 2 days) my own floor plan app, written by Claude and me in Rust and Prolog. It is absolutely great. I plan on selling it for a one-time charge of $50. It is as snappy as any app could be. You can draw multi-floor homes, with each floor having numerous intermediate levels, vast options, and prints in different page formats. Exports to PDF,

I am very experienced in drawing floor plans with Chief Architect and Home Designer Pro and creating very detailed plans.

This app will, of course, be more like CADCraft. But also with much of my touch:

1. For appraisers.
2. ANSI Z765 standard compliance and checking for the US.
3. Other standards for other countries.
4. 12 languages
5. Prolog rule engine to analyze plans for accuracy, compliance, and functional utility.
6. Nice interface with dark/light mode.
etc.

I'll probably work on it for another week (a break from other things I am doing), then put it out for beta testing for registered users with appraisal licenses only.
 
I started an open source group and basically it was crickets despite all the "AI vibe coding for appraisal" noise on social networks. I think in reality, most are not open to collaboration and instead are just trying to sell training, not real solutions. When I reached out to some specific people who made a lot of noise, they all the sudden were not interested...


You've got to frigging do everything yourself. As soon as you ask one person to drag along with you, you have a host of new problems and delays. So, in this respect, Claude - despite its problems is really your best partner. Best and probably ONLY partner + other AI systems if you want to spend resources dealing with them. But people? ... no - they will just slow you down and cause
problems.

Also, if you are working on your own, you will not have much time for support, because you are not charging enough for that. You will also need automated support systems.

This is where it helps to be a master of many skills...
 
I started an open source group and basically it was crickets despite all the "AI vibe coding for appraisal" noise on social networks. I think in reality, most are not open to collaboration and instead are just trying to sell training, not real solutions. When I reached out to some specific people who made a lot of noise, they all the sudden were not interested...

I don't see it that way at all - at least for myself. I'm happy to share anything I've built. What is discouraging for me is that I don't understand github...
 
If you look at the End User Agreement, it is not just that you can't copy or "reverse engineer" their code; you can't steal the design. However, much of the design is just common-sense logic. So where do you draw the line between common-sense logic and creative insight? That is hard to defend against.

Also, a lot of people are just fed up with what Adobe is doing. Their software isn't any better really than it was 10 years ago, except for some AI features. Yet they are charging a fortune to churn out the same old, same old. --- JUST LIKE YOUR APPRAISAL SOFTWARE COMPANIES. It's largely BS - the SAME **** as 15-20 years ago. And they want $1500+ per year for what?

The GSEs handed out detailed specs on UAD 3.6. I had Claude read their specs in and generate code. It did that in several days. Claude said then (around the beginning of 2026) it needed another 4 months to finish - with just one person - myself - helping. The real problem appraisal software companies have producing a UAD 3.6 version is simple incompetence on their part.
 
I keep thinking about this. I've already got a Rust logic engine that spits out valid UAD 3.6 Mismo with the zipped up xml and photos that I use in my mobile app (InspectRadar). I wonder if I should open source it and combine with this "craft" style approach and build a full-on cross report writer UI on top of it. But one that is extendable (think, appraisers can build their own analysis tools and stuff, hook it up to 3rd party data, their own llms, etc). Maybe build in an "app store" so people can give away or sell their own addons. But it would be a new UI interface, not a clone of the WYSIWYG and use best practices for data input. Anyone got some Grok, Claude or OpenAI credits to burn and game for collaborating?

What exactly are you trying to do?

The GSEs handed out detailed specs on UAD 3.6. I had Claude read their specs in and generate code. It did that in several days. Claude said then (around the beginning of 2026) it needed another 4 months to finish - with just one person - myself - helping.

Opus 5.5 wrote me a usable GP form in the 3.6 style in under 15 minutes.
 
I started an open source group and basically it was crickets despite all the "AI vibe coding for appraisal" noise on social networks. I think in reality, most are not open to collaboration and instead are just trying to sell training, not real solutions. When I reached out to some specific people who made a lot of noise, they all the sudden were not interested...


missed this..

I followed but still extremely clumsy in github. I responded to your 3.6 post and asked have you looked into submission costs which is what I read was the massive hurdle. also, love to colab but I already have zero time Im pretty much done with my sites and have some spare tokens what you thinking
 
I keep thinking about this. I've already got a Rust logic engine that spits out valid UAD 3.6 Mismo with the zipped up xml and photos that I use in my mobile app (InspectRadar). I wonder if I should open source it and combine with this "craft" style approach and build a full-on cross report writer UI on top of it. But one that is extendable (think, appraisers can build their own analysis tools and stuff, hook it up to 3rd party data, their own llms, etc). Maybe build in an "app store" so people can give away or sell their own addons. But it would be a new UI interface, not a clone of the WYSIWYG and use best practices for data input. Anyone got some Grok, Claude or OpenAI credits to burn and game for collaborating?
It's possible. I have an html app that replaces all the vendor software which produces mismo 2.6 and 3.6 files successfully. They open in other appraisal apps but since I haven't had a 3.6 order yet, I haven't had an opportunity for a fire test. Large chunks of it were produced ArtCraft style. I plan on porting it to rust at some point.
 
This is how I picture the future appraiser/valuation engineer: They will have their own "Zillow" or AVM or "mass appraisal" system that covers their own appraisal territory and a bit beyond. They will doctor it. They can basically, on the spur of the moment, give you a value for about any residential or commercial property in their city and/or county, and maybe beyond.

They will not be limited to "one property at a time."

Still, some appraisers will be more competent than others, and some less so.

A good appraiser will be able to tell you not only what a property is worth - with a relatively high degree of certainty- but will know so much more:

1. How solid the ground is under it.
2. How solid the structure is and how long it is likely to last into the future with the right kind of maintenance.
3. Its defects, its strengths.
4. Its current functional utility for different user (read: occupant/owner) groups.
5. Its likely functional utility going into the future. For example, does it have movable walls - can it be reconfigured for different occupants? What is its energy profile?
6. Is it a 50-year, 100-year, 200-year, or 300-year home?
7. What kind of catastrophic events should it be able to withstand, and to what extent?
8. What does the maintenance profile look like over the next 10,30,50,80 years?
...

We are moving into a different world, - very fast.

-------------- And

9. Good appraisers will have business connections with contractors and engineers they may work closely with to give advice.

10. There will be a firewall between lenders and appraisers. I think, ideally, some kind of "neutered" government agency will sit between the lender and the appraiser - possibly a highly monitored AI system, built on agreed-upon and strictly enforced rules.
 
Last edited:
Find a Real Estate Appraiser - Enter Zip Code

Copyright © 2000-, AppraisersForum.com, All Rights Reserved
AppraisersForum.com is proudly hosted by the folks at
AppraiserSites.com
Back
Top