When I was talking standards, I wasn't just talking of softawre development - Everything.
"ISO Formats" are possibly a good idea. They are already established standards, and would not involve re-inventing the wheel. I can't instantly think of any that conflict with another - but I'm not overly conversant on every ISO document. There are one or two.
To address 9001 in particular, I personaly think most "features" are inherent or implicit in the "problem identified, we crowd-source solution, improve solution over time" model. Previous experience of such things suggests remarkable effectiveness. With accepting any input for code, only sane code is likely to be selected, but some sort of formal standards on things like comments, formatting etc would make a more uniform experience - like other standards, it's up to the "hobbyist" if they want to adhere to them or not. Those having an interest in applicable fields tend to seek out ISO's for, RFC's etc if only to gain more information.
The state of Asgardian software development will look a lot like Level 1 depicted in your link for some time, by my prediction(based on presented evidence) however, I like to think our number features enough appropriate skill to refine any unwise submissions. But don't forget, chaos too can be a system. It seems to do a good enough job of imposing impression of order on the universe... Over time this will naturally improve. We will have educational capabilites(technically, we could pull that off just with the forum) to assist with such notions and "experts" in the appropriate areas will take up residence for additional guidance, clarification, assistance etc from there being a self-feedback loop that will grow itself past Level 5 that quickly you won't see Level 4. Which isn't much of an issue, we have little need of user-based metrics to track "productivity" - They isn't on payroll, and who cares who does what? what matters is it's done, and done well.
The four points you list would exist in part via the current model. Forum topics effectively can suggest projects, instead of "meetings" there's chaotic posting - but that's like many conversations in a single room.. The textual nature means any features planned are recorded - it's a case of picking through the jumble and assembling it into a tangible proclaimation. ID features, argue about that some. ID ways of providing these features, argue about that some more, From that mess extract all of purpose which aligns intent, maybe argue about that some more - then it's possibly beyond the four points listed and actually building it.
To use a real-world example of the open source car - a common chassis design(based on Smart/Mercedees A class), but everything else is modular. End user can snap together a car almost like lego, with equal flexibility. Several large auto firms are working collaboratively to ensure the individual parts "all fit". To do that, they are working to common standards(and working together to define new standards).
This sort of mentality is essential in software. We need to model our framework(Currently, it's looking like Drupal, That can work, tho) within which we can bolt on our modular parts. To do that, we need to be building code to common standards to ensure all the parts fit, and don't have "unintended consequences". Like the car, start with the chassis. Get the framework well mapped, then provide instructions and guidelines for use, commonly things largely take care of themselves from that point. Any parts fit the chassis and can be selected at will.
If they sensibly use our open sourced materials as base(when we actually have some), they should be somewhat complient at least up to their additions.
To address "core systems", just as an indentification excersize not to consider interactions etc, in order of priority, we need:
*Collaborational tools - from here we can build. Everything. Typical document(text, spreadsheet, database, pictures, .pdf etc) multiuser real-time editing. Incrimental backups of content. These can further be embedded into other services. I'm sure we can spare at least 1TB of HD to holding our output(for now).
*Ticketing systems - this will allow to track issues, across more than just software developments.
*Secure voting system(prequisites passports, in my model at least)
To build something that represents what you intended, it is first sensible to address what it is you would like it to do. After fully identifying our intended features(for now, it would suffice to concentrate on collab»tickets) it would be sane to review how they are intended to interact. Then how they are not intended to interact. From that guidelines can be drafted, argued, adjusted, published. Care at all stages should be made to retain modular flexibility, and promote sane handling of data. Long term stability is mission critical. Taking such into account should allow to define standards to which must be attained/maintained in order for anything submitted to proceed further, such things kind of build themselves in the same way physics shapes the universe(or, more accurately, the universe shapes physics) if approached in the right manner.
For collaborational tools, we would require a storage pool within which to host Asgardian generated documents, but more critically we need guidelines and Accepable use Policy for it's effective, and productive use. Ideally, user access would require to be seperately attributable to the forum, maybe a flag in appropriate(service, not citizen) DB to indicate access revoked due to abuse - lest failure to adhere to useage policy is deemed worthy of isolation from total services. A combination of existing open source softwares, libre office/collabora, git, racktables(not concerned with "productivity" directly but can be abused for much - We will need an asset tracking system eventually anyway)
For ticketing software, Open Source solutions such as osTicket should be trivial to deploy and integrate into existing systems providing an easy to use location for tracking issues and resolutions across infrastructure, not just digital.
If we're really cunning, we can wrangle collaborational tools and a ticketing system into a single proposal