happy you got it solved
As for the high fee... Indeed, there are a lot of unconfirmed tx's being broadcasted lately. This causes some sort of fee-bidding war (since transactions with the highest fee per byte of tx data usually get picked by the miners). There are times when the fees are a lot lower than they are right now tough .

Technically, you can estimate your fee this way:
tx size =~ nr of inputs*147 + nr of outputs*34 + 10 + number of inputs

In your case:
tx size =~ (2 x 147) + (1 x 34) + 10 + 2 = 340 bytes

https://bitcoinfees.21.co/ teaches us the optimal fee at this time is 450 satoshi's/byte

So:
340 * 450 = 153.000 satoshi's = 0.00153 BTC =~ $6.4 (at current preev rate).
However, this is like the optimal fee for which you have a 95% chance of your tx ending up in one of the next 3 blocks... The fee you added should be more than sufficient if you're willing to wait a couple of hours.

No problem, everybody needs to learn... I've been on bitcointalk for several years, but i still learn stuff on a daily basis (usually by saying something dumb, and then getting corrected by Danny Hamilton, Timelord, achow101 or many other longtime members).

In your case:

short answer:  yes, if you do nothing, there's a reasonable chance you'll be able to re-create your transaction in +3 days.

long answer: make sure nobody is re-broadcasting your transaction, and usually electrum will "forget" about your unconfirmed transaction as soon as the electrum server/node you're connected to drops the unconfirmed transaction from its mempool (so if the electrum server you're connected to chanced their default behaviour and keeps the tx in their mempool for 14 days, you'll either need to change servers or wait 14 days)... However, you might need to re-start electrum, or even restore it from seed (worst case scenario).

I don't quite get your question... The price it wat it is, just supply and demand.
If you're just interested in converting your FIAT to BTC, it shouldn't matter if you get 1 BTC for $4000 or 10 BTC for $4000... If you're going to spend this $4000 worth of BTC, you're going to be able to buy products that have a value of ~$4000, irregardless of the amount of BTC you need to transfer.

I own a trezor, a ledger nano S and a ledger HW.1.
The biggest difference between them (IMHO) is the price

The trezor supports a little bit more wallets, the ledger supports a little bit more altcoins... I like the native wallet from the ledger a little bit more than the native wallet of the trezor, altough the fact that you can sync the labels of your native trezor wallet is defenatly a plus.

It seems trezor is releasing new firmware more often than ledger (altough this can just be a gut feeling).

Another difference is entering the pin code:
- A ledger HW.1 will just let you enter the pincode with your numeric keypad
- A ledger Nano S is a bit harder, since you need to use the two buttons on the side to form your pin (which can be a little bit timeconsuming with 8 number pincodes)
- a trezor uses a random numeric keypad shown on its screen in combination with a blank keypad displayed on your screen... It is missing the '0' tough.

I think it all boils down to which coins you want to manage using your hardware wallet... If it's just BTC, you can pick both... If it's BTC and a certain type of altcoin, you might need to buy either the ledger or the trezor.
Also, the trezor ships from the Czech Republic (IIRC) ,the ledger ships from France, so it might save on the shipping time if you're close to one of these countries.

Just to ease your mind: there is no permanent limbo
The way unconfirmed transactions work is pretty easy... You generate a transaction and broadcast it to the nodes you are connected to. The nodes look at the transaction, and if it's new and OK, they store it in their mempool.
A mempool is basically what the word says: an in-memory pool of unconfirmed transactions.
Because so many transactions get broadcasted, the default setting on most nodes is to remove unconfirmed transactions from their mempool after 3 days. Secondly: the nodes allocate a maximum amount of memory to their mempool: as soon as its full, transactions will be removed. Also, when a node reboots, its mempool is usually emptied (at least, this used to be the case, don't know if nowadays the mempool gets saved before a shutdown).

So, as long as nobody rebroadcasts your transaction, a big part of the nodes will have forgotten about your transaction after ~3 days.

Offcourse other nodes (like mine), keep your transaction for 10 days.... But most of the network won't.

This is not a 100% foolproof plan tough... It's possible one miner chose to dedicate 0.1 Tb of memory to their mempool and decided to keep transactions in their mempool for 2 years. So when the average fee drops, this miner might find your 1 year old, unconfirmed transaction suddenly very interesting and decides to add it to the block he's currently working on... But this is just a theoretical example to illustrate my point...

I think that'll be your best option... But look at the bright side, if you succeed in exporting your private keys, you're home free... From that point, you can pick about any decent wallet and use it to sign messages

you don't need to sign up for viabtc. They have a free service over here:
https://pool.viabtc.com/tools/txaccelerator/

They only accept 100 transactions per hour tough, so it's important you enter your tx id + the captcha at XX:59 and hit the submit button at XX:00 because at XX:03 all 100 slots are usually filled.

first of all: it might be a good idear to update your wallets from time to time
With electrum, as long as you have the seed phrase (for a hd wallet) or the private keys you imported originally, not much can go wrong...

As for your question: i don't really follow what's happening at your end, but, generally speaking, you can import private keys from a paper wallet into as many desktop wallets as you want. As long as you don't spend the unspent outputs by creating a transaction and signing it using the private key from your paper wallet, each of those desktop wallets has the potential to spend the unspend outputs (wow, that's a confusing sentence   )

In the end, it might be more usefull to use viabtc's txaccelerator, and just wait untill they hit their next block... Your transaction had a 58 sat/byte fee, so if you submit it to viabtc exactly when the next hour begins, they should accept your transaction, and it should be confirmed in a couple of hours

Like i said: i don't have a lot of experience with multibit, but you can just startup multibit and try it out
I know multibit HD did have the functionality to sign messages, i don't really remembered how it worked and if it was limited tough...

your old bitcointalk account? To such an account, you don't need to sign a transaction, you need to sign a message with an addresses you used in an old, unedited post.. Signing a transaction != signing a message. But to assure you: yes, you can sign messages as soon as you imported your private key into a decent wallet.

This transaction uses 2 unspent outputs as input, and generates one output. I don't think it's double spending any inputs, since in this case, blockchain.info usually gives a warning, and the double spending transaction usually doesn't propagate properly. But like other members already concluded: your only problem seems to be the low fee... (altough i didn't take the effort of properly examining the transaction, at first sight, it looks fine if you disregard the fee)

Just some technical background: there is no "default" fee... If you use an (old) wallet that still uses a "default" fee, for your own sake, please ditch it asap.

On average, one block will be found every 10 minutes (the difficulty adjustment tries to make sure the time between blocks is ~10 minutes, so if more hashrate enters the network, the diff will go up to have this number).
A block has a maximum size of 1 Mb.
So, the number of transactions that can end up in a block is limited to 6 Mb/hour (on average). This is why the fee is calculated based on the size of the transaction. Mind you: size has nothing to do with the amount of BTC that is being transferred.

What does influence the transaction size then? Simple: the number of inputs used and the number of outputs generated by the transaction.
A transaction using 1 inputs of 10 BTC to generate one output of 9.999 BTC will be much smaller than a transaction using 100 inputs of 0.001 BTC to generate 2 outputs of 4.9999 BTC.

Why would you do that?

I don't have any experience with multibit (and only a very limited amount of experience with multibit HD), however, it's worth a shot...
As long as you keep a clean copy of your wallet in a safe place, and use a copy of this copy to place in the multibit appdata folder, it could do no harm (AFAIK). If it doesn't work, you still have the clean copy in the safe place to try decoding your wallet.

That part is easy... Once you have your private key, you can import it into a range of wallets. If you have diskspace and time, you can install bitcoin core, either bitcoind or bitcoin-qt

for bitcoind => "bitcoin-cli importprivkey [yourprivatekey]
for bitcoin-qt => go to the debug console => importprivkey [yourprivatekey]

An other option is electrum (just create a new wallet, but don't generate a new seed, instead chose to generate a wallet using private key(s), then enter the private key when asked...)

there are a lot more options (like  knots, armory, btcd + wallet,...).

And for your last question: yes, after you've imported the private key, most wallets will allow you to sign messages

Indeed, you added 50 sat/byte. That's way to low.
You can try to prioritise the transaction using the viabtc txaccelerator: https://pool.viabtc.com/tools/txaccelerator/ .

A different problem with this tx is that it seems to contain dust:
https://www.blocktrail.com/BTC/tx/98048d83cf5f674bba3d17265190c609ac5874abeed3c7453a5507513425cf47

In the future, always enable opt-in RBF, it's an easy way to bump your fee (in essence, create a "clean" double spend of your unspent outputs in a new transaction that leaves ahigher fee).

Other options might be to do a CPFP or actually double spending the unspent outputs manually (altough such transactions usually don't propagate very well)

Personally, i would keep rebroadcasting it from as many nodes as i could find, hoping that sooner or later, a miner picks it up and puts it in the block he's currently working on.

If you do nothing, chances are that more and more nodes that did know about your transaction will start removing it from their mempool, in long enough time, it'll be like the transaction never existed... This does not happen if you keep rebroadcasting.

It does exist on blockchain.info, it does not exist on a lot of other nodes tough...

EDIT: i tried pushing it to my own node:


Since it worked, i don't have a double spending tx in my mempool (at least, not anymore, no way of knowing if there ever was one).

Strange stuff, the fee is on the low side, but not extremely low... However, i can't find the transaction in the mempool of my node. This usually means that it's a double spending transaction, or it was broadcasted more than 3 days ago...

Same thing for the block explorers: blockr.io doesn't know about your transaction, while blockchain.info seems to have it in it's mempool...

Maybe you can start by re-broadcasting the raw, signed transaction?

If you have a copy of your wallet, you can try to read the directions on this page:
https://github.com/Multibit-Legacy/multibit/wiki/Export%20and%20limited%20import%20of%20private%20keys