[alancalverd]

You need to distinguish between "memory", the combination of hardware and software that permits storage and recall of data, and "memories" - the data within that system. There is no doubt that babies contain the hardware and bootstrap software, but are pretty much devoid of content.

So we move on to the nature of memory. In the realm of artificial memory we use a combination of static and dynamic memory hardware, and various forms of compression and regeneration software. It is arguable that we can segment the workings of the human brain in the same way.

The development of the hippocampus in London taxi drivers has been noted and implies that longterm memory is to some extent a hardware/static element. At the other end of the scale most people can remember a 7-digit phone number for long enough to dial it, but I get the impression from watching the development of pilots' radio communication that eight or nine digits is quite a feat until you acquire some "compression software" by experience: all civilian VHF frequencies begin with 1 and although communication frequencies are allocated in 8.33 kHz steps, the display is deliberately "fudged" to the nearest 5 kHz, sea level barometric pressure is usually between 900 and 1300 kPa, transponder codes are 4-digit octal (0-7 only)....so your longterm memory says you don't need to keep the whole instruction in your short-term memory to execute it.   

Most people can sing back one line of a jazz song (usually 8 or 12 bars, sometimes 16) and that is the essence of leading a crowd chorus, but to recall a whole verse or an entire 3 minute number needs lots of repetition for a beginner. An experienced musician however seems to develop some compression-decompression mechanism that allows adequate recall of an arrangement you may have only played once, years ago.  Note "adequate": like a video image, something can get lost  in the CODEC process, but the joy of live performance is to fill in the gaps!   

So my simplistic analogy is that short-term memory, like dynamic RAM, needs to be refreshed or compressed and transferred to something more akin to static RAM. To pursue the analogy to an extreme, interruption of the refresh process would destroy the contents of dynamic RAM but static RAM decays much more slowly. Evidence? Many people with head trauma never recall the previous two minutes or so, but gradually reconfigure the stuff about names and addresses. Thus we might conclude that longterm memory involves fairly permanent chemical changes that in principle could be detected in a nonfunctional brain, even if we have no idea how to decode them.

Considering how "memory techniques" work, I guess a lot of the CODEC business is done by association and probability, which may account for witness statements often being plausible but contradictory. It also explains the difference between learned language, where we consciously study the formalised structure of a foreign tongue, and language acquired by immersion. The former often gives us a good "passive vocabulary", particularly if we can recognise similarities or have studied a common root language (like Latin) and can read a newspaper or follow a conversation,  but "active vocabulary", the ability to respond  in Klingon without compiling the sentence in English and translating it, is much harder to acquire after the age of five.

Problem is that unless you know a lot about the history and experiences of your dead brain, you won't be able to disentangle the hard molecular data that has been compressed and compiled by association. 

[Bored chemist]

The more complicated and "embedded in the cognitive bits of the brain" memory turns out to be, the more likely it is that death erases it.

[hamdani yusuf]

The distributed nature of the brain lead me to think that you need essentially all of it to decode any of it.

Experiments with brain cancer patients showed that memories can be lost partially. People with amnesia may lose some of memory while retain the others.

Amnesia definition, loss of a large block of interrelated memories; complete or partial loss of memory caused by brain injury, shock, etc.
https://www.dictionary.com/browse/amnesia

Like many systems, some extent of modularity can make them more robust and less vulnerable to disruptions.

[hamdani yusuf]

A baby has life, but no memory.

It comes from the observation that children and adults don't remember events they experienced when they were babies. It doesn't remove the possibility that babies may have short term memories. They might just get lost when they grow up, and the memory space is overwritten by newer memory.
Babies may remember faces, or song, and relate them to events they are experiencing, such as feeling of comfort and warmth brought by their parents.

To me that suggests that the requirement for "life" is easier to meet than the requirement for "memory".
If I'm right then it's likely that any disruption to the brain which extinguished "life" would also destroy "memory".

Both are fuzzy concepts, but useful in decision making. CRISPR mechanism can be thought of as a primitive form of memory.

[Zer0]

The color of my eyes, my skin tone & possibly the nose that sits on my face...All bear resemblance to my Ancestors.

Can these things be also classified as Memory?

Ps - i know I'm going astray, this OP isn't about Genetics, but aren't Seeds just a bundle of memories?

[evan_au]

A baby has life, but no memory.

Studies have shown that even before birth, babies are picking up sounds from their mother, including the sounds of their "mother tongue".

This has been shown to affect the sounds that babies can distinguish after birth.

This is more in the speech processing part of the brain, rather than episodic memories, but I still count it as memory - it is still interconnections between neurons.

aren't Seeds just a bundle of memories?

We are certainly able to decode DNA sequences from plants, animals, ancient humans and even viruses, to an extent that was impossible in the year 1990, when the Human Genome Project was just getting underway.
- Progress during that project was exponential, with the first years devoted to improving techniques, and developing a high-level gene map that later work could fill in.
- As I understand it, after 7 years, when the planned 15-year project timetable was half used, only about 3% of the human genome had actually been decoded
- Some described it as a total waste of time, and should be abandoned
- But in fact, it was declared "finished" (for publicity purposes) 1 year ahead of schedule
- Exponential growth catches everyone by surprise!

Perhaps there will be a similar exponential progress in brain research?

See: https://en.wikipedia.org/wiki/Human_Genome_Project

[Bored chemist]

aren't Seeds just a bundle of memories?

Seeds are mainly food.

[Halc]

The color of my eyes, my skin tone & possibly the nose that sits on my face...All bear resemblance to my Ancestors.

Can these things be also classified as Memory?

It's classified as information. All memory is information, but if you classify all information as memory, then there seems to be no point to having two distinct words to describe the same thing. So it seems that memory is a specific kind of information, and whether a seed constitutes memory or not depends on where you draw the line between your definitions.

So again, a dead person's memories might be gone, but the information possibly isn't, at least not right away. But there seems to be no simple way to read this information, just as you can't easily tell from a DNA molecule what sort of thing is described by it. A good biologist can get pretty close though, at least if it's something related to a life form that he's worked with before.

[Zer0]

aren't Seeds just a bundle of memories?

Seeds are mainly food.

You certainly aren't interested in the Art of Gardening, are You?

It's quite pleasant you know.
Helps to keep oneself calm & composed & in rhythm with nature.

Ps - Try it!

[Bored chemist]

aren't Seeds just a bundle of memories?

Seeds are mainly food.

You certainly aren't interested in the Art of Gardening, are You?

It's quite pleasant you know.
Helps to keep oneself calm & composed & in rhythm with nature.

Ps - Try it!

You seem to have muddled what they are made from with what they are for.

[hamdani yusuf]

So it seems that memory is a specific kind of information, and whether a seed constitutes memory or not depends on where you draw the line between your definitions.

Memory usually refers to information acquired from experience. Biology borrows the word and use it to describe similar traits for species other than humans.

Electronic engineering and computer science extend the scope further to any information which can be easily read/recalled.
But the op seems to restrict the scope to human memory.