'storage bins stacked vertically'
Prior art ...
http://www.montypython.net/scripts/society.php
AutoStore, a Norwegian manufacturer of robotic warehouse storage and retrieval systems, says it is suing UK online retailer and tech outfit Ocado. It is claimed the Ocado Smart Platform – which Ocado has sold to US retailer Kroger as well as British chains Marks & Spencer, and Morrisons – unlawfully infringes its patented …
The complaints don't appear to be listed yet in the public-facing court filing systems. We're seeking out copies, and once we've got the patent list, we'll add it to the story.
Edit: Complaint and patents added -- it's got little to do with tape drives unless you squint really hard.
C.
I wonder if the robotic tape drive was mentioned because its so similar to robotic tape drives like StorageTek Automated Cartridge System. It is very possible for two people to come up with the same idea based on another idea (Robotic Tape Drive) that end up very similar. Stuart Lee does a bit about this in his Plagiarists Corner like
LEE & HERRING / THE GOVERNMENT – EXAM MARKING SYSTEM SUMMER 2020
:)
And he's also pointed out issues with his own stand-up with stuff that pre-dates his to show he's come up with an idea without knowing the original existed.
These are quite different to a tape library. Usually there is one tape arm which is attached to the frame.
These warehouse systems use autonomous robots on a grid of tracks. They move quite fast and there is some sort of orchestration to (mostly) avoid collisions. The robots are like 100+Kg.
Also below the robots there are stacks of bins. I have never seen a tape library with more than one tape in a slot.
IMO there is plenty of scope for patentable algorithms and mechanics.
Multi-mechanism tape libraries are definitely a thing - start with a small number of tapes and a drive, add more and more tapes, eventually you need more i/o to make use of the capacity, especially when it's being used as proper near-line storage rather than just writing tracts of backups with the occasional restore. Also allows for ready duplication if you want to export a bunch of tapes while keeping the same set locally. But of course the arms aren't autonomous, no need to track state; much like a hard disk, the smarts to pull which tapes/data to which drives at what times come from the controller or the host computers.
If they did a good job, you wouldn't know which one you got back. An acquaintance sent some mechanical parts to china for copying. Replicated items pretty decent copies - original brand name, part number, patent numbers all crisply cast/etched in place just like the originals.
Got to say we did that once, we borrowed a CCTV swan neck wall mount off the manufacturer purely so we could measure the thread to get a local metal worker to duplicate it, then we sent one of the copies back.
We realised our mistake a couple of days later when we found the original on the warehouse shelf.
Remarkably the manufacturer either never noticed, or didn't care we'd sent a copy back
How much patent overlap is too much?
I mean, these systems look pretty similar in the photos, but general automated storage has been around for decades. We have similar systems at work but it's rotated 90 degrees to the side. The bins travel along platforms and are inserted into vertical stacks. To me the function is no different and we have had those systems for 25 years.
Using that logic I could design a car that was still a functional car but looked slightly different, then sue anyone who copied it.
What I was reminded of - I think I saw it on TV once:
https://en.wikipedia.org/wiki/Chinese_typewriter
1970s model - There are thousands of symbols, so you just steer the picker over the field of type pieces till you find the one that you want, and you type it. At least, I think that's how it goes.
I suppose there are more details in the warehouse overhead railway for robots thing, and if it is true that Ocado took a good look at the other mob's setup and THEN built their own, then that is a bit whiffy.
Could write packing code worth a damn. In my experience they ignore instructions not to pack meat with fresh food (e.g. fruit), put heavy items on top of deformable items (like bottles on top of bread), don’t pack items prone to leakage separately (liquid soap bottles for example), and are completely irrational about what they do pack in their specialty sealed packs (boxes of cat food?).
Back in the 18th century, the RSA (the Society for the Advancement of Arts, Manufactures and Commerce at the time - they got their Royal Warrant a lot later) established a system of awards to reward inventors who were prepared to share their ideas for the advancement of all, rather than protect them with patents for commercial greed. It's a pity that system is no longer running.
In my view, a patent should be worth no more than the financial investment that went into its development. For example, if it took 10 hours for an engineer to draw it up, them 10x the audited hourly cost to the company (for simplicity, say, $100/hr). If another company wants to use the idea, they pay $1000 up front; same for anyone else. Similarly, if a development cost $20m, each company wanting to use it pays $20m up front. The organisation patenting the idea should be getting profit from its own use - if not, they'll get their investment returned by the competitors. Determining the cost shouldn't be any harder than writing the patent, especially as any smart company will be writing development costs off against tax (and subject to audit and criminal sanctions if over-estimated). It should also help kill off the patent trolls as they can only charge what they paid for the patent, which would be no more than the previously declared cost. It also stops simple patents being used to stifle the application of innovation - which was a driver for the RSA 250 years ago.
Very much an over-simplification of what is needed and plenty of folk will come up with reasons why it can't work - but that shouldn't stop efforts to improve it.
PS If this idea is introduced, I claim this post as prior art so nobody can patent the new patent system!
I'd have to disagree.
Patents exist to protect intellectual property, because unlike physical property it can be stolen without difficulty and without evidence (e.g. the property actually missing).
So it is necessary to have a system where you prove you had the idea first and then protect it from anyone else, at least for a short period of time.
For example, it can take tens of thousands of models, thousands of lab tests, dozens of human studies and years of FCC approval to get just one drug on to the market. As a result, the first pill off the conveyor belt may have cost 100 million to produce. However, the second pill will only cost 20 cents, since it's just a small bunch of chemicals.
If you just let anyone else make the 2nd pill, no one will ever afford to be able to make the first pill.
So do we need patents? Yes.
Are there people who abuse the system? Sadly also yes. Nothing is perfect (and some patent systems are less perfect than others).
I know nothing about this case, but if AutoStore came up with something inventive and patented it, and Ocado now use that without AutoStore's consent in a jurisdiction covered by their patents, then AutoStore may well have a case. Ocado, meanwhile, can try to show that the patent's aren't valid.
It's messy because people and business are messy, not because of the patent system itself.
> The organisation patenting the idea should be getting profit from its own use - if not, they'll get their investment returned by the competitors.
This excludes every University then. A Uni may have a great idea for improving a warehouse but they aren't logistics organisations themselves so they won't get any direct benefit. The "excess profits" from obtaining a patent and licensing the idea pays for the research spent on duds that never make it.