Re: Lessons (was Re: Objects and Relations)
Date: 13 Feb 2007 02:33:00 -0800
Message-ID: <1171362780.457121.283800_at_v33g2000cwv.googlegroups.com>
On Feb 12, 6:27 pm, Bob Badour <bbad..._at_pei.sympatico.ca> wrote:
> Bob Badour wrote:
> > What lessons have you learned that you might want to relay? (Please
> > enumerate.)
Heck, why not. With the all-encompassing caveat that the follwoing is all off the top of my head:
- Implementation != Theory
- To produce good IT we should be analysing _Information_ not just Technology.
- Value based addressing.is superior to OID's.
- Values should never be hidden.
- Propositions are simply true or false and do not 'change'.
- OID's break liebniz equality.
- Encapsulation has a negative impact on shared data, leading to
query bias.
- Query neutrality is important for good management of shared data.
- Do not confuse conceptual/physical/logical layers.
- Purely Navigational Databases are inferior to Declarative Databases.
- Nulls are both nonsensical and generate logical errors.
- There are two types of updates masquerading as one - domain redefinitions/proposition replacement,
- RM redefines some mathematical concepts - tuple/relation/crossproduct.
- Semi-structured data has no definition.
- Keys are not entity identifiers, nor vice versa. Keys are the antecedents of material implication.
- Surrogates are still attributes, just unobserved ones.
- Combining XML with xlink and xpath to create a data model is like grafting arms and legs on to a hamburger.
>
> 1. The self-proclaimed seldom are.
> 2. Not every human mind is equally able.
> 3. Intellectual honesty is a prerequisite to learning.
> 4. Not every human mind is capable of abstraction.
> 5. Some non-human minds are capable of experimentation and methodology.
> 6. Success is context-sensitive.
> 7. Our tools affect our minds.
> 8. The predicate calculus is more illuminating than the set algebra.
> 9. Programmers are drawn more to the algebra than the calculus.
> 10. Theoretically non-updatable views should nevertheless be updatable.
> 11. It is very difficult to respond coherently to that which is incoherent.
> 12. Good intentions can lead to disaster.
> 13. Selfish motives can lead to great good.
> 14. Continuous assessment and improvement are best.
> 15. Empiricism is the only hope to understand reality.
> 16. Caution is appropriate when designing.
> 17. Wanton recklessness is appropriate when imagining.
> 18. The scientific method keeps natural human deficiencies in check.
> 19. We should strive to tackle problems at or near the limit of our
> capability.
> 20. Austere mental discipline is required for real progress.
> 21. A pleasant demeanor can make nastiness palatable.
> 22. An unpleasant demeanor can unmask intellectual dishonesty.
> 23. Some with an unpleasant demeanor are simply nasty.
> 24. There is no stopping the invincibly ignorant.
> 25. Humans tend toward irrational and non-rational belief.
> 26. Status hierarchies are very important to human happiness.
>
> > What audiences do you try to reach with these lessons?
>
> 1. Family
> 2. Friends
> 3. The readers of c.d.t
>
> > What approaches have you used?
>
> 1. Profanity
> 2. Intimidation
> 3. Disdain
> 4. Understatement
> 5. Overstatement
> 6. Provocation
> 7. Asking questions
> 8. Absense
> 9. Sophistry
> 10. Referral to better sources
> 11. Self study
> 12. Appeal to reason
> 13. Logic
>
> > What other approaches are you aware of?
>
> 1. Absolute empiricism with quiet skepticism
>
> > What other audiences might exist?
>
> 1. Colleagues
> 2. Students
> 3. Web surfers
> 4. Other programmers and data managers
> 5. Other types of professionals
> 6. Executives
> 7. Academics
> 8. Random strangers
>
> > What big questions remain unanswered in your mind?
- Is a fully relational language tractable.
- Is it possible to generalize Codd's insights without relations.
- Is an MV system (ignoring current suggestions and implementations) ever justifiable.
- View updatability.
- Is it a valid approach to consider a stuctural layer between logical and physical.
- Is it possible to provide a mechanism to prevent the same mistakes continually occurring in IT.
>
> 1. What are the biggest challenges to distributed optimization?
> 2. Where will the next great insights lead us?
> 3. How will germline genetic engineering affect our offspring?
> 4. How long will it take to achieve engineered negligible senescence?
> 5. Where can I best devote my effort?
>
> > How do you measure success?
>
> 1. When I gain an insight
> 2. When I communicate an insight
> 3. When someone exceeds my expectations
> 4. When I see something admirable
> 5. When I improve an efficiency
> 6. When I remove a systemic defect
> 7. When it compiles
> 8. When it behaves as expected in the debugger
> 9. When it runs
> 10. When my dogs run and play
> 11. When I increase my serenity
- Insight
- Elegance
- Serenity